RCAC - RCAC News

Purdue research computing grid is no Toy Story, but a new animated video about it is close

Thu, 12 Nov 2009

If Purdue University computer graphics technology student Micah Bojrab gets his dream job with a Hollywood animation studio, a system for harnessing unused computers to do cutting-edge research will have played a part. The DiaGrid system developed by Purdue inspired a video combining computer animation and real-life scenes that would make Pixar proud.

See the DiaGrid video at: www.youtube.com/watch?v=CH_YHGYQl2g.

“This is the most complex thing I’ve ever done,” said Bojrab, a Purdue master’s student who graduates next spring. Bojrab led the team that created the animation, which took more than 750 hours to render. The team also included Purdue students Chris Sprunger and Pan Wen.

The project gave the students real-world experience beyond anything possible in the classroom, said David Braun, who heads the visualization facility operated by ITaP, Purdue’s central information technology organization. ITaP, which also operates DiaGrid, employed the students for the video project. Jason Doty, a producer/director for ITaP’s video and multimedia unit, shot the live footage and combined it with the animation.

DiaGrid pools computers over the Purdue campus network and off campus via the Internet and fast research networks. Whenever machines in the pool are idle—at night, when their owners are at lunch and otherwise—the system sends work to them. Campus Technology Magazine selected DiaGrid for a 2009 international Campus Technology Innovators Award. Purdue is promoting DiaGrid, and premiering the DiaGrid video, at SC09, the world’s largest high-performance computing conference, which opens Monday in Portland, Ore., and ends Friday Nov. 20.

The video mixes animated computer chips, stamped with the DiaGrid logo, arrayed like an extremely long row of dominoes. When the inevitable happens, and the chips begin to fall, they travel from Purdue’s supercomputing center into the world at large. The dominoes move through Purdue campus scenes, past high-tech windmills in central Indiana and down rustic railroad tracks, among other things. The line also rolls by scenery on the campuses of Purdue’s DiaGrid partners, including Notre Dame’s “Touchdown Jesus.” In the end, the domino chips cover a map of the U.S. forming DiaGrid’s logo.

The video highlights two key elements of DiaGrid: the nearly 30,000 computer processors the system makes available to researchers nationwide and Purdue’s goal of a national partnership to push the total over 100,000. Current partners include Indiana University, Indiana State University, the University of Notre Dame, the University of Louisville, the University of Wisconsin, Purdue’s Calumet and North Central campuses, and Indiana University-Purdue University Fort Wayne. DiaGrid has been used for everything from imaging viruses at an atomic level to probing the formation of the Solar System.

While high-tech computer graphics are central to the video, clever low-tech tools enabled finishing touches. Doty couldn’t find a good sound effect for the falling domino chips, until he hit on the idea of recording two CD jewel cases clacking together. Braun found the reflective chrome ball, used to capture lighting information at the live scenes so the animation could be realistically lighted—for $11 in the aisle devoted to garden ornaments at a local home improvement store.

Writer: Greg Kline, science and technology writer, Information Technology at Purdue (ITaP), 765-494-8167 (office), 765-426-8545 (mobile), gkline@purdue.edu

Sources: Micah Bojrab, 260-402-2476, mbojrab@purdue.edu

David Braun, 765-496-7888, info@envision.purdue.edu

Jason Doty, 765 -494-7091, rjdoty@purdue.edu

Purdue highlights cloud computing tools at SC09 supercomputing conference

Wed, 11 Nov 2009

Facebook for scientists—but built to facilitate serious research rather than socializing—and an award-winning method for putting idle computers to work on scientific breakthroughs are Purdue-developed technologies in the spotlight at the SC09, the world’s largest high-performance computing conference. Purdue is highlighting the HUBzero and DiaGrid technologies in the University’s booth at SC09, which runs through Friday Nov. 20 in Portland, Ore.

HUBzero is a unique, soon-to-be open source software platform developed by Purdue for deploying and applying computational research tools, visualizing and analyzing results interactively and publishing them, all through a familiar Web browser. Built-in social networking features akin to Facebook create communities of researchers and educators in science, engineering, medicine, almost any field or subject matter.

DiaGrid works by pooling computers over the Purdue campus network and off campus via the Internet and fast research networks. Whenever machines in the pool are idle—at night, when their owners are at lunch and otherwise—the system sends work to them. Campus Technology Magazine selected DiaGrid for a 2009 international Campus Technology Innovators Award.

Purdue has an automated system to link the computers of SC09 participants to the pool during the conference. The Purdue booth includes a scoreboard to keep track of whose machines are running the most jobs.

The booth is designed to promote Purdue, ITaP, the University’s central information technology organization, and the Rosen Center for Advanced Computing, ITaP’s research and discovery arm. ITaP technologists developed HUBzero and DiaGrid.

“DiaGrid and HUBzero are model technologies for enabling research that Purdue is making available to the world,” says John Campbell, associate vice president in charge of research computing for ITaP, who heads the Rosen Center. “As the premier conference for research computing, SC09 is a prime place to showcase these technologies.”

Purdue’s booth also provides information on other Purdue research initiatives, academic information to potential Purdue students and information to job seekers about positions with Purdue, ITaP and the Rosen Center. Nearly 10,000 people attended the conference in 2008.

Purdue has become a recognized leader in “cyberinfrastructure” with the development of HUBzero, which powers nanoHUB.org and many other Web-based “hubs” for research collaboration. NanoHUB is an international resource for nanotechnology theory, simulation and education with tens of thousands of users.

“Like no other platform, HUBzero can host interactive simulation tools, so users aren't just reading about research, they can experience it,” says Michael McLennan, senior research scientist and hub technology architect at Purdue. “HUBzero allows users to work together as they interact with content.”

Other hubs link researchers transforming laboratory discoveries into new medical treatments; working to revolutionize cancer prevention, detection, treatment and care delivery; promoting assistive technologies innovation to better serve those with disabilities; advancing manufacturing techniques; and engineering earthquake-resistant buildings, bridges and related structures. Purdue is now working in a consortium with Indiana University, Clemson University and the University of Wisconsin to advance the technology even further.

A hub will be at the center of the Network for Earthquake Engineering Simulation (NEES), a $105 Million National Science Foundation program announced in September, which is led by Purdue. Purdue electrical and computer engineering Professor Rudolf Eigenmann, co-principal investigator of NEES, is giving a workshop titled “Cyberinfrastructure for Earthquake Engineering” at the Purdue booth.

McLennan is hosting two workshops on HUBzero and one about nanoHUB during SC09. Purdue scientist Mathieu Luisier is offering a workshop on using petascale supercomputers to simulate nanoscale electronic devices for the next generation of electronics, a central focus of nanoHUB.

DiaGrid includes computers in student computer labs, offices, server rooms and supercomputing clusters and is the first multi-campus collaboration of its kind. Purdue’s partners in DiaGrid are Indiana University, Indiana State University, the University of Notre Dame, the University of Louisville, the University of Wisconsin, Purdue’s Calumet and North Central campuses, and Indiana University-Purdue University Fort Wayne.

Together, they now make nearly 30,000 processors available for research jobs ranging from understanding the Solar System’s formation to imaging the structure of viruses at near-atomic resolutions in an effort to develop new ways of battling viral illnesses, from swine flu and the common cold to West Nile virus and AIDS.

“The sheer size and ingenuity of the initiative, as well as the diversity of computing resources represented in the grid, really set the project apart,” Geoffrey Fletcher, editorial director of Campus Technology, said in announcing the Campus Technology Innovators Award for DiaGrid.

The schedule of presentations in the Purdue booth is:

* Tuesday, 11 a.m., “HUBzero: Cyberinfrastructure for Research, Education, and Collaboration,” Michael McLennan, Purdue senior research scientist and hub technology architect.

* Tuesday, 1 p.m., “Towards Petascale Simulations of Nanoelectronics Devices,” Mathieu Luisier, Purdue and Network for Computational Nanotechnology research scientist.

* Tuesday. 2 p.m., TeraGrid science gateways for earth science modeling and data management, Carol Song and Lan Zhao, Purdue TeraGrid team.

* Tuesday, 3:10 p.m., “Cyberinfrastructure for Earthquake Engineering,” Rudolf Eigenmann, Purdue electrical and computer engineering professor and co-principal investigator of the Network for Earthquake Engineering Simulation.

* Wednesday, 11 a.m., “NanoHub: An International Resource for Nanotechnology Theory, Simulation and Education,” Michael McLennan, Purdue senior research scientist and hub technology architect.

* Wednesday, 1 p.m., TeraGrid science gateways for earth science modeling and data management, Carol Song and Lan Zhao, Purdue TeraGrid team.

* Wednesday, 2 p.m., “HUBzero: Cyberinfrastructure for Research, Education, and Collaboration,” Michael McLennan, Purdue senior research scientist and hub technology architect.

More information:

www.youtube.com/watch?v=CH_YHGYQl2g

www.dia-grid.org

hubzero.org

Writer: Greg Kline, science and technology writer, Information Technology at Purdue, 765-494-8167 (office), 765-426-8545 (cell), gkline@purdue.edu

Sources: John Campbell, 765-494-1289, john-campbell@purdue.edu

Michael McLennan, 765-494-6495, mmclennan@purdue.edu

Cluster Challenge: Purdue, ITaP student supercomputing team ready for SC09

Wed, 11 Nov 2009

Purdue students Alex Younts and David King helped build Purdue’s latest supercomputer in July. Call it a summer workout. Younts and King are now part of a Purdue student team running its own supercomputer in a competition at the SC09 Conference in Portland, Ore., which runs through Friday Nov. 20.

The 2009 Cluster Challenge entry from Purdue is like a mini version of the Coates cluster ITaP, Purdue’s central information technology organization, installed over the summer. Coates is the fastest campus supercomputer in the Big Ten and one of the fastest in the world. The student cluster is made up of similar HP computers containing AMD microprocessors. It also boasts similar high-speed networking, called 10 Gigabit Ethernet (10 GbE), from Juniper Networks to link the individual computers into a single, powerhouse machine. One challenge for the team has been tuning the software it must run in the competition to take advantage of the unusually fast connection, says Jeffrey Evans, an associate professor in the Department of Electrical and Computer Engineering Technology who’s working with the team members.

“I think that helped us get some of the pieces together more quickly,” Younts, a junior in computer science from West Lafayette, Ind., says of the experience assembling Coates. Like King, a senior in electrical and computer engineering technology from Lafayette, Ind., Younts is a student worker for ITaP, which operates Coates for Purdue faculty researchers. Hundreds of volunteers and ITaP employees built the new supercomputer in just a morning on July 21.

The refrigerator-sized Cluster Challenge machine, which ITaP helps sponsor, doesn’t have Coates’ more than 10,000 processors. But the student cluster still has more than three times the horsepower of the machine Purdue used for the competition when it entered the first time in 2007, Younts and King note. They think that may be enough to give Purdue an edge at SC09.

The Cluster Challenge, which draws teams of students worldwide, is part of the largest international conference for high-performance computing, networking, storage and analysis. Cluster Challenge teams—limited to undergraduates—have to run a set battery of benchmarking programs and working scientific applications as efficiently as possible. The student supercomputers churn through simulated data provided by the competition organizers, with little likelihood of actually finishing and a goal of simply getting as far as possible in the time allotted. The student team members share the load monitoring and adjusting their machines around the clock over three days as they work to achieve maximum performance.

The 2009 applications cover climate modeling, computational chemistry and data visualization, among other things. The students on the Purdue team spent most of the semester setting up and tweaking their machine, according to Professor Evans and Preston Smith, a senior systems administrator at the Rosen Center for Advance Computing, ITaP’s research computing arm. Smith serves as team leader.

Most of the students participate in Evans’ high-performance computing class at Purdue as part of their involvement in the Cluster Challenge. But they spend up to 20 hours a week on the effort overall and have to juggle other classes, homework and exams to attend SC09 for the competition.

The other team members are Max Hapner, a senior in computer and information technology from South Whitley, Ind.; Alex Miller, a sophomore in earth and atmospheric sciences from Zionsville, Ind.; Michael Niksa, a senior in computer engineering from Cedarburg, Wis.; Charles Timko, a junior in computer science from Highland, Ind.; and Michael Wleklinski, a junior in chemistry and statistics from South Bend, Ind. Hapner and Niksa also worked on the Coates cluster.

Besides Evans’ class, the experience has other educational benefits. For example, Wleklinski joined to learn about the chemistry software NWChem, one of the 2009 Cluster Challenge applications, which he’s interested in using in his own research.

“I'd like to attend graduate school and hopefully get a PhD in either analytical or physical chemistry with a master’s in statistics and then enter industry,” Wleklinski says. “By participating in this I've gained skills in computation that have allowed me to expand my horizons.”

Meanwhile King, who will graduate next spring, plans to take advantage of another Cluster Challenge benefit: the SC09 jobs fair “Take lots of resumes, you may have a job offer before you leave,” Evans advises the students. “That’s how highly Cluster Challenge team members are regarded.”

Miller was attracted by the opportunity to work with WRF a weather modeling and prediction application. “I wanted to get experience in modeling,” he says. “I also thought it would help to get involved with networking outside of my department. This experience will help with my grad school ‘resume.’ In terms of a career, the cluster challenge has helped give me an idea of what is involved in working with prediction systems on a daily basis.”

Hapner wants to become a professional system administrator eventually. He figures a key part of advancement will be knowing at least something about how most things in a data center work. He says the Cluster Challenge was a good opportunity to expand his knowledge.

Niksa is primarily interested in “parallel” computing to get software to take full advantage of multiple processors now featured even in laptops. “It seems as if that's the direction most branches of the discipline will be heading in the future if they're not on that track already,” he says. “I believe working with the parallel technologies required for the Cluster Challenge project put me on the cutting edge of software and computer engineering.”

Timko works for ITaP as a research application programmer and already had an interest in working with supercomputing clusters and parallel programming when Younts encouraged him to join the Cluster Challenge team. He’s been developing a version of the open source Ubuntu Linux operating system for use on clusters. But he thinks he might make a career of developing operating systems for a much smaller kind of computer—smart phones.

More information:

www.itap.purdue.edu/newsroom/detail.cfm?NewsID=1949

www.purdue.edu/uns/photos

Writer: Greg Kline, science and technology writer Information Technology at Purdue (ITAP), 765-494-8167 (office), 765-426-8545 (mobile), gkline@purdue.edu

A training course covering Vizard 3.0 will be offered at 11:30 a.m. Nov. 17 and Dec. 1 by ITaP’s data visualization facility the Envision Center for Data Perceptualization.

The sessions will be held in the Envision Center, located in Stewart Center B001 off the hallway between the Stewart Center and the Purdue Memorial Union. The course is aimed at participants with possible applications in psychology, liberal arts, entertainment, architecture, computer-aided design (CAD), scientific visualization, virtual training environments, and many others.

Vizard is a virtual reality programming and development platform that allows novice to expert users to easily create 3-D environments, computer controlled avatars and interactive activities.
The first session will cover the basics of working in Vizard, including object creation, manipulation and selection. The second session will include more advanced programming techniques allowing participants to start experimenting with their own virtual environments.

For more information contact David Braun at info@envision.purdue.edu.

Participants who wish to bring a laptop with Vizard installed to follow along during the course should see www.worldviz.com to download a 90-day trial version.

Study gives a clearer picture of how land-use changes affect the U.S. climate

Wed, 11 Nov 2009

Researchers say regional surface temperatures can be affected by land use, suggesting that local and regional strategies, such as creating green spaces and buffer zones in and around urban areas, could be a tool in addressing climate change.

A study by researchers from Purdue Universtiy and the universities of Colorado and Maryland concluded that greener land cover contributes to cooler temperatures, and almost any other change leads to warmer termperatures.

Read more...

Purdue Google Gadgets offer satelitte data visualization for everybody

Fri, 16 Oct 2009

Purdue collects masses of satellite data for use by scientists and state and local officials, but anybody who uses Google now can share in some of this data cornucopia.

Two Google Gadgets -- a type of small Web application -- created by Purdue staff and students working for ITaP let users see a visual, animated rendering of cloud cover, cloud pressure and water vapor as they change over the entire country and individual states, along with wildfire threats in the Midwest, among other things.

Research suggests urban sprawl, wet falls and winter affect severe weather

Fri, 16 Oct 2009

Previously rare big city storms -- like a tornado August 19 that downed trees and ripped off roofs in downtown Minneapolis and the powerful thunderstaorms in New York City a day earlier -- may not be so unusual anymore.

As large urban areas continue to expand, they appear to influence tornadoes and other severe weather, research suggests. Cities could be even more at risk if located in a region experiencing a wet fall or winter, according to researchers from Purdue University and the University of Georgia.

Event will highlight research computing services available to Purdue researchers

Fri, 28 Aug 2009

Research computing services and cyberinfrastructure available to Purdue researchers in all fields will be the focus of Cyberinfrastructure Day, a series of informational sessions scheduled for Sept. 30.

Aimed at faculty, research scientists and graduate students, the event, including lunch, will be held from 8 a.m. to 5 p.m. in the Stewart Center. Registration is required. To register visit www.rcac.purdue.edu/events/cidays.cfm. There is no fee.

The program will include hands-on sessions and case studies designed to assist current and potential users in accessing Purdue’s research computing resources and cyberinfrastructure. The keynote speaker is Mark Lundstrom, director of the Network for Computational Nanotechnology and Don and Carol Scifres Distinguished Professor of Electrical and Computer Engineering.

Topics scheduled to be covered include ITaP’s community cluster and storage programs and Condor distributed computing pool and Purdue’s partnership through ITaP in the National Science Foundation’s extensive TeraGrid cyberinfrastructure for research and education.

The program also will cover ITaP services for providing data visualization assistance, high-resolution satellite and remote sensing data, collaboration on grant applications, and more. In addition, the event will highlight Purdue-developed HUBzero, a Web-based platform that makes it easy for researchers to connect with colleagues throughout the world and share ideas, tools, computational resources and data storage while satisfying funding agency grant requirements.

The event is sponsored by ITaP and ITaP’s research and discovery arm the Rosen Center for Advanced Computing. The Rosen Center’s staff of computing professionals and computer scientists specializes in supporting research and discovery through high-performance computing. The center provides access to leading-edge computational and data storage systems, as well as expertise in a broad range of high-end computing and data management activities.

Center staff:

· Evaluate, deploy, and support hardware and software for numerical- and data-intensive research computing;

· Promote the effective use of computing systems and application software through training and education, consultation, and documentation;

· Contribute to the discovery process through the development of effective computing techniques and data capabilities;

· Partner with researchers during the development of grant proposals by offering expertise in the assessment of hardware and software requirements.

· Assist researchers in obtaining resource allocations and access to national cyberinfrastructure such as the TeraGrid and the Open Science Grid.

Writer: Greg Kline, science and technology writer, Information Technology at Purdue (ITaP), 765-494-8167, gkline@purdue.edu

Consortium offers Purdue researchers access to massive computational power

Fri, 28 Aug 2009

Purdue researchers will have a chance on Sept. 23 to learn about opportunities for research computing and computing education projects related to the Blue Waters petascale computer being built at the National Center for Supercomputing Applications (NCSA).

A Purdue seminar on Blue Waters will be held at 10 a.m. in room 121 of the Burton D. Morgan Center for Entrepreneurship, 1201 West State Street in Purdue’s Discovery Park. There will be a question and answer session at 10:45 a.m. Registration is not required. The event is sponsored by ITaP and ITaP’s research and discovery arm the Rosen Center for Advanced Computing. For more information please see http://www.rcac.purdue.edu/events/bluewaters.cfm.

The featured speaker will be William T.C. Kramer, deputy project director and co-principal investigator for Blue Waters at NCSA. Kramer earned his bachelor’s and master’s degrees in computer science from Purdue.

Funded by the National Science Foundation, Blue Waters is expected to be the most powerful supercomputer available for open scientific research when it comes online in 2011. The project includes the NCSA, the University of Illinois, IBM and the Great Lakes Consortium. Purdue is one of the consortium’s 28 members. Gerry McCartney, Purdue vice president for information technology and chief information officer, serves as a consortium board member.

A key element of the Blue Waters project, the consortium of universities, colleges, national research laboratories and other institutions is designed to foster use of petascale computing, among other things through development of new software, applications and technologies. The group’s educational and workforce development program is aimed at making sure advances are passed to the next generation of researchers and applied to frontier questions in science, technology, engineering and the social sciences.

Writer: Greg Kline, science and technology writer, Information Technology at Purdue (ITaP), 765-494-8167, gkline@purdue.edu

ITaP units' virtualization partnership avoids duplication, better serves researchers

Mon, 03 Aug 2009

Recent growth in demand for computational resources from Purdue researchers prompted ITaP to expand its high-performance computing hardware drastically, including the Steele community cluster built in 2008, one of the world’s Top 500 supercomputers, and the Coates cluster built July 21.

The growth also has strained capacity to handle routine tasks, such as managing password authentication, print serving, or routing email messages about research jobs running on ITaP’s supercomputers. The Rosen Center for Advanced Computing, the research and discovery arm of ITaP, uses regular server computers for that type of work, and for less demanding computing needs of Purdue researchers it serves, such as developing Web-based science tools.

For example, the Purdue Ionomics Information Management System (PiiMS) is a Web portal aimed at better understanding how plants take up, transport and store nutrients and toxics—knowledge which could benefit both human health and the natural environment. PiiMS makes curated data on many thousands of plants freely available to the public via the Web. Now, Rosen Center services supporting the PiiMS portal can take advantage of the same robust virtualization infrastructure as other University functions.

To meet the broader needs of the research community, the Rosen Center is using the “enterprise class” virtualization already in place for Purdue’s academic and business computing needs. In effect, teams within ITaP have decided against developing a second virtualization infrastructure solely for the University's research mission—an option the Rosen Center considered—and agreed to leverage their varied expertise to more effectively serve faculty researchers and other users, said John Campbell, associate vice president in charge of ITaP’s research computing efforts.

Michael Rubesch, ITaP’s systems and operations head, said ITaP can provide cost-effective service for other campus units that would benefit from virtualization as well. This also will allow them to avoid duplication of efforts within Purdue as the Rosen Center has done.

Purdue was recognized by InfoWorld magazine in 2007 for its virtualization efforts. The magazine listed ITaP’s program among its Top 100 innovative solutions to information technology challenges developed by corporations and institutions.

Virtualization allows a server computer to function is if it were multiple servers, even multiple servers with different operating systems. Users generally can’t tell whether they’re running applications on a virtual machine or on hardware.

Purdue’s virtualization program initially allowed it to move the work of 140 servers to three machines. That move reduced power usage by 84 percent, cut the amount of space taken up by servers to a tenth and saved $325,000 a year. Today, ITaP runs nearly 300 virtual machines on just eight physical servers. The system handles everything from certain student data to sales at the Purdue golf courses—and now includes support services for supercomputing.

The Rosen Center’s virtual machines run on the same servers ITaP systems and operations uses for nonresearch computing purposes. The virtualization platform employs virtual machine software from VMware (a subsidiary of EMC), ITaP’s partner. Rosen Center staff members can create, configure and administer the virtual machines from across campus, while ITaP systems and operations continues to handle the underlying infrastructure and offers seasoned technical support on virtualization.

Rubesch said ITaP initially got into virtualization to address power and cooling costs and space limitations at Purdue’s primary data center in Freehafer Hall. Similarly, an increased demand for power and cooling for larger research systems has led the Rosen Center toward virtualization.

The disaster recovery aspects of virtualization also were attractive. Virtual machines running on a server that goes off line can be moved to another host machine on a network in the VMware architecture. It also takes less time to create a virtual machine, just minutes, than to provision equivalent hardware, Rubesch said.

Disaster recovery was a factor in the Rosen Center’s move to virtualization. However, pressed for capacity in its Math Building data center, the savings in space, power and cooling was even more important. The move also keeps Rosen Center staff—along with its computers—focused on the center’s main function, research computing.

Writer: Greg Kline, science and technology writer, Information Technology at Purdue (ITaP), 765-494-8167, gkline@purdue.edu

New Purdue cluster was working on the TeraGrid before it was even finished

Mon, 03 Aug 2009

Purdue University’s new high-performance computing cluster, named Coates, was available on the TeraGrid even before staff and volunteers finished building it July 21. The nodes in the new cluster, made up of more than 700 HP dual quad-core computers with AMD processors, joined Purdue’s Condor pool almost as soon as they were racked and connected, part of an automated configuration process. Purdue makes the Condor pool available as a TeraGrid high-throughput computing resource. It has been used for everything from imaging the structure of viruses at an atomic level to simulating the formation of the Solar System.

Coates is a “community cluster” paid for by a cooperative of Purdue faculty members and campus departments and operated by Information Technology at Purdue (ITaP), Purdue’s central information technology organization. Investors in the cluster always have access to the nodes they purchase and potentially more when others’ nodes are idle. That still leaves idle time, however, which is made available through the Condor pool to local users, the TeraGrid and the Open Science Grid. In this way, Purdue has been able to keep its Steele cluster built in May 2008 working more than 90 percent of the time.

The addition of Coates brings the number of processors in the Condor pool to approximately 27,000. Coates is expected to reach its 1,280-node capacity in the fall of 2009, which will bring the Condor pool to nearly 33,000 cores. Coates should rank in the top 50 on the November 2009 Top 500 Supercomputer Sites list. It is the first internationally ranked academic supercomputer to be solely wired with 10 Gigabit Ethernet.

Purdue’s Condor pool is at the center of DiaGrid, a multi-campus Condor pool with Indiana University, Indiana State University, the University of Notre Dame, the University of Louisville, the University of Wisconsin, Purdue’s Calumet and North Central campuses, and Indiana University-Purdue University Fort Wayne. Purdue is seeking other partners with a goal of growing DiaGrid to 100,000 processors.

More information:

www.rcac.purdue.edu/userinfo/resources/coates

www.itap.purdue.edu/newsroom/detail.cfm?NewsID=1949

Writer: Greg Kline, science and technology writer, Information Technology at Purdue (ITaP), 765-494-8167, gkline@purdue.edu

Source: Carol Song, 765-496-7467, carolxsong@purdue.edu

Purdue partnership for putting idle computers to work on research wins international award

Mon, 03 Aug 2009

Purdue is being recognized as a worldwide campus technology innovator for its DiaGrid system harnessing what would be otherwise wasted computing power for major research projects.

Campus Technology Magazine has selected DiaGrid for one of its 2009 Campus Technology Innovators Awards.

ITaP interns get real experience with research and technology, graduate student life

Thu, 23 Jul 2009

ITaP summer research intern Lutalo Webb already knew he would be back when he arrived at Purdue. The May engineering graduate from Vanderbilt University had applied and been accepted for graduate school in West Lafayette. But the time he's spent analyzing tests of advanced optics technology promises to give him a running start, and solidified his choice of major professors.

Encore to ITaP's first supercomputer barn-raising is even better than the original

Tue, 21 Jul 2009

Last year ITaP staff and a small army of volunteers had the Steele supercomputer built by noon. The anticipation was that the new Coates cluster would take longer.

Coates is a larger project and it required installing special network cards. Then, the hardware used to mount and group the individual computers in the refrigerator-sized racks that form the supercomputing cluster turned out to require some adjustment, too.

So Coates was finished by, well, around 11 a.m.

Credit people like Nathan Heck, an ITaP security engineer who installed 66 of the network cards over a couple hours. Dwight McKay, director of systems engineering for ITaP research computing, said based on pre-build testing he figured the installers would be able to do 12 cards an hour.

“Concentration and motivation,” Heck quipped when asked for his secret.

Heck and other motivated workers did everything from unboxing the computers—and sorting all the packaging for recycling—to shuttling them to the basement machine room in the Mathematical Sciences Building, sliding them into the racks and plugging in the cabling to tie them all together into one big machine.

Nearly 300 volunteers from campus units around Purdue signed up to participate, along with out-of-town visitors from Indiana University, the University of Iowa, Michigan and Michigan State. They worked with the kind of dispatch one might expect from a crew stacking sandbags to ward off a looming flood. The flurry of activity drew a stream of distinguished sightseers, including Purdue President France A. Córdova.

“It’s amazing how many people volunteered to get this all together,” said Andrea Leydet, an ITaP research computing intern who will be a senior in computer information technology at Purdue Calumet.

Leydet and Sara Lopez, also an ITaP summer intern, helped sort materials for recycling.

“I’m pretty excited,” said Lopez, a Purdue Calumet junior. “It’s my first time ever getting to do anything with a cluster. It’s a new experience.”

The experience was an old one for Matt Link, research computing systems director at Indiana. He helped build Steele in May 2008.

“I was cleaning up trash downstairs last year,” quipped Link, who helped make adjustments to the rack mounts for Coates. “I’ve moved up.”

Jerry Protheroe, Iowa campus technology’s research services manager, said he and colleague Jeffrey DeReus made the trip to get tips for a supercomputer project Iowa has in progress—and to help out as “payback” for assistance from ITaP’s John Campbell and Bill Whitson in applying Purdue’s “community cluster” model.

By midmorning Randy Herban of ITaP research computing could point to a graph on a laptop used to monitor the new system and announce that Coates was running research jobs already.

“This is live data,” Herban said pointing to a growing cascade of blue peaks moving across the screen. “Stuff is actually starting to happen.”

McKay joked that ITaP might just have to fabricate the computer chips for its next supercomputer along with installing the network cards. “Maybe that will slow us down,” he said.

Coates, like Steele, is a community cluster, paid for by faculty on campus who contributed research money to fund the purchase. The supercomputer is named for Clarence L. “Ben” Coates, head of Purdue's School of Electrical Engineering (now Electrical and Computer Engineering) from 1973 to 1983 and a driving force behind high-performance computing at Purdue.

Pictures of the day's activities may be found at http://www.purdue.edu/uns/photos/

Writer: Greg Kline, science and technology writer, Information Technology at Purdue, 765-494-8167, gkline@purdue.edu

ITaP seeking volunteers to help build Purdue's next supercomputer

Wed, 01 Jul 2009

Think of the fun of unboxing a new computer, plugging in the keyboard, mouse and monitor and finding out what the speed demon, relative to last year’s model, can do. Now, ITaP is looking for volunteers who want to multiply their new computer pleasure by hundreds of times.

Purdue’s central information technology organization will hold an installation day for the University’s latest research supercomputer from 8 to 4 p.m. July 21. People who want to help build the Coates community cluster should visit www.rcac.purdue.edu/cores/ to register. Click on the “Sign up to volunteer for Installation Day” link.

On installation day, volunteers will assemble at the welcome tent in front of Felix Haas Hall, behind the Elliott Hall of Music and off University Street between Second and Third streets. HP, which won the bidding to supply the computers for Coates, will provide lunch.

A volunteer meeting will be held July 17 to explain the plan for installation day.

ITaP needs people with experience installing rack-mounted computers, hardware such as expansion cards inside machines and cabling. But general help unpacking and moving equipment, recycling packaging materials and cleaning up also is needed.

Last year, more than 250 staff members and volunteers assembled the new “Steele” cluster. Some volunteers even came from rival Indiana University, attracted by the idea of building an 18 wheeler-sized supercomputer in a day, a process that normally takes weeks. The high-tech barn raising generated national media attention.

Coates is designed to exceed Steele in capability, with up to 1,280 separate rack-mounted computers linked together to create one big supercomputer and fast 10 Gigabit Ethernet networking. The new cluster is being named for Clarence L. “Ben” Coates, a driving force behind the high-performance computing and networking plan that led to the creation of the Engineering Computer Network for Purdue’s engineering schools.

Through “community clustering,” ITaP pools funds from grants, faculty startup packages and institutional sources to make more computing power available than faculty and campus units could afford individually for major engineering, science and social science research projects. ITaP installs, administers and maintains those systems, including security, so researchers can concentrate on doing research not on running a high-performance computing system.

One goal of community clustering is maximizing the use of resources by sharing computing power researchers use only part of the time with their peers, who can make use of it during what might otherwise be idle time. Researchers always have access to the computing power they purchase, and potentially more if they need it. The combining of orders also attracts a better price from vendors. Purdue saved more than a half million dollars on Steele.

Writer: Greg Kline, ITaP science and technology writer, 49-48167, gkline@purdue.edu

Virtual Omaha Beach: Purdue team recreates D-Day battlefield, launches learning environment where information searches for user

Thu, 28 May 2009

Purdue and Indiana University researchers are commemorating the 65th anniversary of D-Day by releasing the first version of a 3-D, interactive model of the Omaha Beach battlefield.

A demonstration will take place at the Advanced Visualization Lab Tuesday on the Indiana University-Purdue University Indianapolis campus. The event will demonstrate the 3-D model of a key battlefield in the allied invasion to liberate France on June 6, 1944, and several methods of using the system for learning.

“The model, which includes 3-D pillboxes, beach obstacles, field guns, or ships is, in effect, a Web interface,” said Purdue Professor Sorin Adam Matei, its creator and leader of Visible Past, a project developing similar virtual historical sites. “By simply pointing to an object or location of the virtual battlefield, you can call up more information, collaborate with other learners, or add new information.

Matei, an associate professor of communications, is an affiliate of the Envision Center for Data Perceptualization, part of Information Technology at Purdue (ITaP) and its Rosen Center for Advanced Computing. Envision Center staff members and students help develop the D-Day and other Visible Past models.

Students using Virtual Omaha can walk or fly through the model of the beach and the French countryside behind it, taking the perspective of the American or German troops who fought during World War II. Users can inspect troop positions from all angles and information about the digital artifacts encountered can be brought up automatically.

“The really novel aspect of the project is that if another group uses the model while you are visiting it, any information that they add to it will become available to you instantly,” Matei said.

Besides the Web, the Visible Past models can be run in 3-D virtual environments like the three-walled, room-sized system at the Envision Center. They also work in Google Earth or through free, open source software for 3-D Web-based modeling.

In the near future, people visiting Omaha Beach in Normandy, France, will be able to use an iPhone application, developed by Matei, to access the model and collaborate remotely with others.

“We will have professors delivering a tour to a group in Normandy, while students in Indiana will be able to see and hear through (an immersive virtual environment) what their colleagues see and do in France,” Matei said.

The iPhone application also can be used as a “location-aware” guide for Omaha Beach or any historical site documented by the Visible Past project. When visiting the Roman Forum, for example, information about the nearest building could be sent automatically by Visible Past to the iPhone. “This is ubiquitous computing, where information searches for you,” Matei said.

Virtual Omaha is one of the more than two dozen 3-D models, including several UNESCO World Heritage sites, that can be used for teaching and collaboration through the Visible Past project. The models, some of which were created collaboratively by a worldwide community of students, scholars and amateur historians, are enhanced with information collected by Purdue students. Among the projects under development are the Roman Forum, the Taj Mahal and the Statue of Liberty.

Directions to Virtual Omaha demonstrations:

To introduce Virtual Omaha to the public in anticipation of the June 6th D-Day anniversary, public presentations will be given at 12:30 and 1:30 p.m. Tuesday, June 2, at the Advanced Visualization Lab on the Indiana University-Purdue University Indianapolis campus. The lab is located on the fourth floor of School of Informatics, 535 W Michigan St., Indianapolis. A Google Map of the event location and directions are available at http://tinyurl.com/omahaevent. Seating is limited. Contact Sorin Matei for a reservation at 317-416-5807 or at smatei@purdue.edu. Please arrive at least 10 minutes early. There are a limited number of parking meters in the lot directly west of the building. Additional parking can be found in the University Conference Center parking garage approximately a quarter mile west on Michigan St. To see the IUPUI campus map for more information visit http://www.iupui.edu/about/maps/campusmap.html. At the School of Informatics building, check in with the receptionist on the first floor and request an escort to the Virtual Omaha event.

Source: Sorin Adam Matei, 317-416-5807, smatei@purdue.edu

Related Web sites:

Visible Past:

http://visiblepast.net/home

Omaha Beach Google Earth Model:

http://www.visiblepast.net/gwiki/images/d/dc/Omaha.kmz

Omaha Beach historical details in Google Earth:

http://visiblepast.net/gwiki/images/d/d4/Explore.kml

(To experience Visible Past as a teaching environment both links should be clicked in succession.)

Omaha Beach, 3-D VRML model (requires plug-in):

http://www.avl.iu.edu/projects/VisiblePast

Sorin Adam Matei’s research blog:

http://matei.org/ithink

ITaP taking orders for new research supercomputing cluster and individual systems

Fri, 15 May 2009

Orders for nodes in the new Coates community cluster are now being taken by Information Technology at Purdue (ITaP).

This also is an opportunity for departments or individuals purchasing computing equipment for use outside the cluster to take advantage of big price breaks available in a group purchase, said John Campbell, associate vice president for information technology, who heads the Rosen Center for Advanced Computing, ITaP’s research and discovery arm.

The Rosen Center plans to build the new cluster this summer. The plan calls for the supercomputer to be running research jobs by July.

“The community cluster program allows faculty to pull resources together into a more significant computational resource,” Campbell said. “Faculty can use their nodes and also borrow unused cycles from other faculty partners.”

Space limitations in ITaP's MATH building machine room mean the Coates cluster can have a maximum capacity of 1280 nodes. New orders won't be taken once that limit is reached. Those ordering equipment for use outside Coates, however, will be able to do so for six months, Campbell said.

For a catalog of available hardware and to order Coates nodes, or to make individual or departmental purchases, visit: http://www.rcac.purdue.edu/order.

Through “community clustering,” ITaP pools funds from grants, faculty startup packages and institutional sources to make more computing power available than faculty and campus units could afford individually for major engineering, science and social science research projects.

The Rosen Center installs, administers and maintains the community cluster systems, including security, so researchers can concentrate on doing research not on running a high-performance computing system.

One goal of community clustering is maximizing the use of resources by sharing computing power researchers use only part of the time with their peers, who can make use of it during what might otherwise be idle time. Researchers always have access to the computing power they purchase, and potentially more if they need it.

The combining of orders during purchasing also attracts a better price from vendors, who are willing to provide significant price breaks on large orders. Purdue saved more than a half million dollars in building the Steele cluster in May 2008. Many of the more than 30 faculty “investors” took the opportunity to add to their orders, ending up with more computing power for the dollars they had planned to spend.

The new cluster is being named for Clarence L. “Ben” Coates. He was a driving force behind the high-performance computing and networking plan that led to the creation of the Engineering Computer Network (ECN) serving all of Purdue’s engineering schools.

Writer: Greg Kline, science and technology writer, Information Technology at Purdue (ITaP), (765) 494-8167

gkline@purdue.edu

Purdue TeraGrid staff and students present at CondorWeek 2009

Tue, 28 Apr 2009

Condor Week is a four-day annual gathering of Condor developers, collaborators and users at the home of the Condor development team: University of Wisconsin, Madison. This is a well-attended forum where researchers and practitioners exchange ideas and experiences, learn about latest research, and influence the research and development directions for the Condor software. Condor Week 2009 took place during April 20th - 23rd, 2009. Purdue TeraGrid staff Preston Smith presented our experience and challenges of building and operating a large scale Condor pool (BoilerGrid) that harvest idle computer cycles from many corners of the Purdue's main campus, Purdue's regional campuses, and a number of academic institutions including University of Notre Dame, Indiana University, and University of Wisconsin, Madison. The total number of CPUs currently in the pool is more than 23,000. Purdue ECE graduate student Jungha Woo, who works on the TeraGrid project, presented the Condor Java Daemon Core project, a collaboration among Purdue, Indiana University, and University of Wisconsin to develop a scalable tool for accessing and managing Condor job status reports in a large pool such as the Purdue Condor pool. Nathan Patterson, a Purdue Mechanical Engineering graduate student presented his research on rounding in the beer distribution game. Patterson's research was made possible with the Purdue Boilergrid.

Smith's presentation: Condor on campus: BoilerGrid, DiaGrid and Beyond (can be accessed at:
http://www.cs.wisc.edu/condor/CondorWeek2009/condor_presentations/smith_cw2009_Purdue.pptx

Woo's presentation: Condor Java Daemon Core
http://www.cs.wisc.edu/condor/CondorWeek2009/condor_presentations/woo-JavaDaemonCore.pdf

Patterson's presentation: A High Throughput Computing Analysis of Rounding in the Beer Distribution Game "http://www.cs.wisc.edu/condor/CondorWeek2009/condor_presentations/patterson-BeerDistributionGame.pptx

Additional presentations at the CondorWeek 2009 can be found at http://www.cs.wisc.edu/condor/CondorWeek2009/.

New TeraGrid Campus Champion making Purdue researchers an

Wed, 15 Apr 2009

A “people person” with a background in helping users take advantage of high-performance computing resources whose job already focuses on assisting researchers in using the TeraGrid—you couldn’t ask for a better description of a good TeraGrid Campus Champion.

That’s just what Purdue is getting in Kim Dillman, a research programmer for Information at Technology at Purdue (ITaP) and ITaP’s Rosen Center for Advanced Computing who was recently named the university’s first Campus Champion.

ITaP offers Purdue researchers fast, reliable, secure data storage

Tue, 07 Apr 2009

Information Technology at Purdue (ITaP) is now offering no-hassle data storage in the same way its community cluster program helps faculty members do research using high-performance computing — that is, without having to worry about running a computer system.

Faculty can purchase trays of highly reliable, high-performance scratch disk space for research use through ITaP’s Rosen Center for Advanced Computing (RCAC). Participants purchase disk media. RCAC provides the infrastructure, five years' warranty support and systems administration.

“This is really intended for people who need high-speed, reliable storage connected to hundreds of clients,” said John Campbell, associate vice president for information technology, who heads the Rosen Center.

For more information including a summary of the storage options and prices, visit: http://www.rcac.purdue.edu/hardwarepurchase/storage.cfm.

In addition, ITaP is preparing to take orders for its next community cluster, to be named Coates, which will be built in the summer of 2009.

The Coates project also is an opportunity for departments or individuals purchasing computing equipment for use outside the cluster to take advantage of price breaks available in a group purchase, Campbell said. The combining of orders during purchasing attracts significantly better prices from vendors.

Through the community cluster program, the Rosen Center, the research and discovery arm of ITaP, pools funds from grants, faculty startup packages and institutional sources to make more computing power available than faculty and campus units could afford individually for major engineering, science and social science research projects. The Rosen Center installs, administers and maintains those systems, including security, allowing researchers to concentrate on doing their research.

For more information about the community cluster program, go to: http://www.rcac.purdue.edu/userinfo/communityclusters.cfm

Writer: Greg Kline, science and technology writer, Information Technology at Purdue, (765) 494-8167, gkline@purdue.edu

Luncheon talk to cover high-performance storage for research data

Fri, 27 Mar 2009

High-performance storage solutions that can meet and adapt to the demands posed by large research data sets with numerous users will be the focus of a luncheon at Purdue Wednesday, April 1, featuring James Reaney of BlueArc Corporation, a leading expert in the field.

The luncheon is scheduled for 11:30 a.m. to 1 p.m. in 1142 Lawson Computer Science Building, corner of Third and University streets. The event is sponsored by BlueArc, which provides storage solutions at Purdue’s Rosen Center for Advanced Computing, the research and discovery arm of Information Technology at Purdue (ITaP). ITaP is co-sponsoring the luncheon.

Attendance is open to all Purdue faculty and staff. Although not required, registration is encouraged to ensure sufficient food for the luncheon. You may register at:

http://www.itap.purdue.edu/training/registration/?offeringid=3677

Researchers and system administrators with large-scale, high-performance storage needs in particular could benefit by attending and learning about BlueArc's technologies and how ITaP is using them.

Reaney is BlueArc’s executive director for strategic technologies and coordinates activities with research customers worldwide. Prior to joining BlueArc, he was network and server operations manager for life sciences at Harvard University. With more than 18 years of experience as an information technology director and computing, networking and storage analyst in various high-performance computing research environments, his experience provides him with a solid working knowledge of research customers' day-to-day operations and needs.

Reaney’s talk will contrast consumer and enterprise storage technology; discuss the

benefits of managed enterprise storage for research needs; and cover research data management, including a cradle-to-grave data lifecycle. He will present a case study on accelerating research application performance for a next-generation DNA sequencing pipeline, a project of the BlueArc Center of Excellence at Purdue in partnership with ITaP.

Gerry McCartney, Purdue vice president for information technology and chief information officer, will give opening remarks at the luncheon.

Writer: Greg Kline, science and technology writer, Information Technology at Purdue (ITaP), 765-494-8167, gkline@purdue.edu.

Patient safety focus of immersive virtual environment for training pharmacists

Wed, 11 Mar 2009

When Tara Holt, a third-year Purdue University pharmacy student from Frankton, Ind., steps into a pharmacy clean room for the first time, she’s likely to experience a little déjà vu.

The room should look and sound familiar. Nothing ought to feel strange about standing encased in a sterile hair cover, mask, gown, gloves and booties. That’s because Holt and her classmates will have experienced it all before—in a virtual version of a pharmacy clean room. The computer-generated, 3-D immersive environment created in a Purdue project brings to mind the holodeck on the Starship Enterprise, for a serious purpose rather than recreation.

Prehistoric global cooling caused by C02, research finds

Tue, 03 Mar 2009

Ice in Antarctica suddenly appeared — in geologic terms — about 35 million years ago. For the previous 100 million years the continent had been essentially ice-free.

The question for science has been, why? What triggered glaciers to form at the South Pole?

Survey asks Purdue researchers what they want in new computer system

Fri, 13 Feb 2009

Purdue researchers are being asked for feedback on the features they need in a new supercomputing cluster planned by Information Technology at Purdue (ITaP) and the Rosen Center for Advanced Computing.

The Rosen Center plans to build the new community cluster, to be named Coates, in the spring of 2009. The project also is an opportunity for departments or individuals purchasing equipment for use outside the cluster to take advantage of price breaks available in a group purchase, said John Campbell, associate vice president for information technology, who heads the Rosen Center.

Purdue, IU consortium aims to drive new era for research, education online

Fri, 13 Feb 2009

A powerful Web platform that allows scientists to share research modeling tools, presentations and experimental findings is the focus of a new partnership between Purdue and Indiana universities.

Purdue is launching a consortium for joint development of the platform, called HUBzero, originally created by Purdue researchers. Gerry McCartney, Purdue's vice president for information technology and chief information officer, will tell the university's trustees on Friday (Feb. 13) that IU will be the consortium's first member.

Purdue research aims to stave off electronics industry crisis

Wed, 14 Jan 2009

The design of microprocessors and other devices central to the electronics age faces a crisis. The 40-year process of transistor downscaling has led to atomic-scale features, making devices subject to unavoidable manufacturing irregularities and to power densities comparable to a nuclear reactor. A new design approach embracing the atomistic, quantum mechanical nature of the constituent materials is necessary to continue developing future generations of more powerful—and lower energy, less hot—computer chips and other electronics along with their components, such as switches.

Working with the Rosen Center for Advanced Computing, Purdue electrical and computer engineering Professor Gerhard Klimeck and his lab are developing tools and techniques that allow such an approach. In the process, the Purdue researchers are winning access to some of the most powerful supercomputers in the world. The Texas Advanced Computing Center, on whose “Ranger” supercomputer (No. 6 on the latest Top 500 list) Klimeck and Mathieu Luisier, a research faculty member with Purdue’s Network for Computational Nanotechnology, have been running their code, is highlighting the work.

Divide and conquer: distributed graphics rendering

Thu, 08 Jan 2009

Rendering frames in a complex animation can tie up workstations for days or weeks. A two-minute animation at 30 frames per second will typically take over 100 hours to render on a single computer. Sharing the load among hundreds or even thousands of machines dramatically reduces a job’s run time—and the time required to complete a project.

Enter the Distributed Rendering Environment, or DRE, developed at Purdue University five years ago for computer graphics students. It draws on the Purdue Condor pool, a system for sharing unused computing time on more than 20,000 linked processors in computers at Purdue and its partner campuses. This pool is part of both the TeraGrid and the Open Science Grid.

Time on giant computer will allow Purdue researchers to look at very small things, lots of them

Thu, 08 Jan 2009

Purdue researchers will study the next generation of computer chips before they are even built using one of the most powerful supercomputers in the world.

The Department of Energy awarded electrical and computer engineering Professor Gerhard Klimeck and colleagues 5 million computer hours under its INCITE (for Innovative and Novel Computational Impact on Theory and Experiment) program. The program, open to all scientific researchers and research organizations, including industry, is for computationally intensive research projects of large scale that can make high-impact scientific advances through a major allocation of computer time, resources and data storage, according to Energy Department Office of Science officials.

Klimeck’s is one of 25 new projects—competitively selected for their technical readiness and scientific merit—that will advance research in key areas such as astrophysics, climate change, new materials, energy production and biology for applications ranging from designing quieter cars and improving commercial aircraft design to developing nanomaterials and simulating earthquakes, said an Energy Department press release announcing the awards Thursday (Dec. 18).

“From understanding the makeup of our universe to protecting the quality of life here on Earth, the computational science now possible using DOE’s supercomputers touches all of our lives,” said DOE Under Secretary for Science Raymond Orbach. “By dedicating time on these supercomputers to carefully selected projects, we are advancing scientific research in ways we could barely envision 10 years ago, improving our national competitiveness.”

Klimeck’s Purdue lab will use its allocation for atomic-level modeling and simulation, in effect virtual nanotechnology engineering, of tiny transistors. The work is designed to address the increasing difficulty of designing microprocessors and other devices central to the electronics age at a time when their components have reached and are dipping into the nanoscale. A nanometer is a billionth of a meter, about 50,000 times smaller than a human hair is round.

Klimeck said the research could help in shifting nanotechnology from the realm of scientific discovery to practical application. “The availability of such large-scale machines and the codes that can utilize them enables us to move nano science to nano engineering,” he said.

The Purdue researchers’ goal is understanding design implications of advanced transistors at the nanometer scale. To do that, they will model and simulate tens of thousands to millions of atoms in the devices using nano-electronic modeling software, OMEN, developed by Purdue research professor Mathieu Luisier, and NEMO3D, developed by Klimeck, associate director for technologies at the Network for Computational Nanotechnology at Purdue.

Capturing the actions and interactions of that many atoms takes massive computing power. With the INCITE award, the Purdue researchers will be able to draw on the Jaguar supercomputer at Oak Ridge National Laboratory in Tennessee, which has a peak performance as of November in excess of a petaflop, that is a quadrillion mathematical calculations per second. It would take the Earth’s entire population, more than 6 billion people, 460 years working with hand calculators to accomplish what the machine can do in a day. The system, No. 2 on the list of the world’s Top 500 supercomputers for science released in November, includes more than 180,000 processors, kind of the brain of a computer. Klimeck said access to Jaguar will allow large-scale simulations the researchers couldn’t perform on other, less powerful machines.

Getting software to make efficient use of all those processors is a challenge, but Klimeck and Luisier already have been able to “scale” OMEN and NEMO3D for thousands of processors, more than 59,000 in the case of OMEN. They use a variety of computational resources, including Purdue’s Steele cluster, which also made the Top 500 supercomputers list. Steele, with more than 7,100 processors, is operated by the Rosen Center for Advanced Computing, the research and discovery arm of Information Technology at Purdue, the University’s central information technology organization.

Writer: Greg Kline, (765) 494-8167, gkline@purdue.edu

Source: Gerhard Klimeck, (765) 494-9212, gekco@purdue.edu

More information:
Versions of NEMO3D and OMEN are available for open public use at the Web-based simulation portal http://www.nanohub.org in end-to-end online tools Quantum Dot Lab, Bandstructure Lab, and OMEN nanowire. Quantum Dot Lab has been employed by more than 1,600 users and Bandstructure Lab by more than 2,000 users, while OMEN nanowire has just been released. In all nanoHUB.org is serving more than 87,000 users annually in 172 countries with online simulation tools, nanotechnology tutorials and more. More information on NEMO3D and OMEN can be found at http://cobweb.ecn.purdue.edu/~gekco.

NanoHUB is based on the Purdue-developed HUBzero technology, http://hubzero.org, an elegant, easy-to-use system for creating online communities where educators and researchers can share ideas, tools, computational resources and data.

Drought information to be more predictive and versatile for a variety of users

Mon, 08 Dec 2008

Odd as it may seem given that a considerable swath of Indiana flooded this spring, parts of the state were almost dry enough in September to start using the “D” word before Hurricane Ike sent a spate of rain north.

Drought, said Dev Niyogi, a Purdue earth and atmospheric sciences and agronomy assistant professor and the Indiana State Climatologist, can sneak up on you slowly and your outlook on it when it arrives may depend on where you live, what you do for a living, or even your weekend plans. One person’s worrisome dry spell can be a baseball fan’s stretch of good weather.

The complexity of drought triggers, conditions and impact has Niyogi and Purdue colleagues, including a team from the Rosen Center for Advanced Computing, working on new way of tailoring drought-related information for a variety of users and on a more localized basis. DRInet, funded by the National Science Foundation, will be a Web-enabled system for accessing data related to the onset of drought and its impacts across disciplines and domains, along with tools to probe, integrate and visualize that data to create new knowledge and aid decision making in near real time.

The system also will make use of the TeraGrid, the world's largest open science computing network, which is funded by the NSF, too. Purdue through the Rosen Center, the research and discovery computing arm of Information Technology at Purdue (ITaP), is a TeraGrid partner and resource provider.

Carol Song, a senior research scientist at the Rosen Center, a principal investigator on the DRInet project and the Purdue TeraGrid leader, leads the cyberinfrastructure development for DRInet.

“The technology piece for this project will ramp up quickly because we are able to use the data and Web technologies developed at Purdue in a number of NSF-funded projects including the TeraGrid and nanoHUB.org,” Song said. “We will extend the software to incorporate drought information management and dissemination, and provide a community Web space to share data, visual products and knowledge to improve understanding and aid decision making.”

Song, Niyogi and colleagues envision a prototype drought knowledge generator resulting from the three-year project that can be used by, among others, climatologists, hydrologists and other researchers alongside farmers, water system managers, businesses, economists, and local, state and national policy makers.

“This is a mix of research, as well as an applied aspect,” said Niyogi, whose own research focuses, in part, on extreme weather events such as storms, floods and droughts.

Niyogi and colleagues said the intent is to expand on, rather than replicate, other drought-related research efforts and the active online community portal U.S. Drought Monitor, which takes more of a national view and tends not to dip below the state level in its analyses.

“(U.S. Drought Monitor) gives you very general information,” said Purdue Professor Indrajeet Chaubey, who is looking at the impact of drought on water quality as part of the DRInet project. “It doesn’t give you the specific information we hope to provide.”

“There are lots of things we want to know,” added Chaubey, an agricultural and biological engineering and earth and atmospheric sciences professor. “The impacts of drought are under-examined now. I think there’s a need for tools that people can use to examine these type of problems.”

Rao Govindaraju, the Christopher B. and Susan S. Burke Professor of Civil Engineering, said resources like U.S. Drought Monitor focus on current conditions—not what could happen in the future given those conditions.

The goal is to make DRInet a more predictive tool, first by “hindcasting,” looking at past records to see what the system might have predicted versus what actually happened, and where it might need to be modified to yield accurate projections, Govindaraju said. DRInet will be something like an enhanced Geographic Information System for drought-related purposes, focused on synthesizing various types of data to gain an understanding of what is happening more than just mapping.

Niyogi said the researchers want to illuminate the broad impacts droughts may have in obvious areas like water supply and crop yields but also on less obvious things like air quality and human health. The picture might incorporate such things as the economic cost and the number of people affected as well.

“The cost of a drought can be more than you incur from a flooding event or a hurricane,” Govindaraju said.

That’s because droughts tend to happen, and have an impact, over much longer periods of time and much larger areas.

“It is an increasing concern, both floods and droughts, the two extremes,” Govindaraju added. “The way the climate system is projected, it seems we will have more frequent and severe events.”

Climate change predictions that anticipate more hydrological extremes, including droughts and floods, are one motivation behind DRInet. The nation’s possible shift from oil to biofuels—derived from crops on which drought can wreak havoc—only raises the stakes, Niyogi said.

Purdue was able to assemble the kind of interdisciplinary team that will be needed to get a picture of the complex web of factors involved in the onset of droughts and drought impacts, not to mention communicate the information to a mix of audiences.

For example, computer science professors Daniel Aliaga and Christoph Hoffmann will work on ways to visualize drought-related data in 2-D and 3-D. The goal is to help researchers make better sense of the data, but also to aid the non-scientific audience DRInet is intended to serve, said Aliaga, who will be probing the data for effects of drought in urban areas, for instance on land use patterns.

Meanwhile, Chaubey and his lab already have been developing mathematical models and high performance computing tools to study the impact of agricultural and other non-point pollution sources on water quality, for entire watersheds and under a variety of weather scenarios and time scales. The techniques can apply to examining drought.

The project also dovetails with existing Purdue efforts in cyberinfrastructure for environmental monitoring, such as the Purdue Environmental Observatory Data Portal, a gateway to environment-related data managed by Purdue that allows researchers and students to access the material without having to learn the underlying software technology. The environmental observatory portal also is a TeraGrid resource.

Other members of the DRInet team include Jacob Carlson, a data research scientist for the Purdue libraries, and Lan Zhao, a Rosen Center research scientist.

Purdue has a number of off-campus partners, such as the National Drought Mitigation Center at the University of Nebraska Lincoln, the National Weather and Climate Center and the National Oceanographic and Atmospheric Administration Central Region. The project also includes a variety of academic partners, including the University of Illinois, Michigan State, Ohio State and Wisconsin, along with Central State University in Wilberforce, Ohio, Western Kentucky and the Indian Institute of Science in Bangalore, India.

Writer: Greg Kline, (765) 494-8167, gkline@purdue.edu

Sources: Dev Niyogi, (765) 494-6574, climate@purdue.edu
Carol Song, (765) 496-7467, carolxsong@purdue.edu
Indrajeet Chaubey, (765) 494-1162, ichaubey@purdue.edu
Rao Govindaraju, (765) 496-3402, govind@purdue.edu
Daniel Aliaga, (765) 496-7943, aliaga@cs.purdue.edu

Purdue project aims to move visual mountains, and quickly

Mon, 08 Dec 2008

In an age of scientific visualization employing huge datasets, and of networked instruments that produce data in torrents, bandwidth is an issue. More of it may be available than ever before and it may be faster than in the past, but the pipeline remains relatively limited and expensive.

“Networking is not cheap,” said Purdue computer science Professor Chris Hoffmann. “Bandwidth is not very big at the moment. We’re working on technology to make all that, shall we say, affordable, particularly the networking.”

A system being developed at Purdue could make moving large, high-resolution scientific visualizations long distance over a network like the Internet manageable and it may open up new opportunities in interactive video conferencing and online collaboration as well, said Hoffmann, who also is director of Purdue’s Rosen Center for Advanced Computing, the research and discovery arm of Information Technology at Purdue (ITaP), the university’s central information technology organization.

The idea is to do the heavy lifting for a visualization on computers close to a big dataset being drawn upon, pack the result for shipping over a network, with no loss of fidelity, and send it to the user’s end. There, it gets unpacked locally, ready to run with fluidity comparable to streaming video—or television—but less consumption of bandwidth along the way.

The “lossless” technique uses features of the data to be visualized, and an understanding of them built into the algorithms behind it, to eliminate redundancy and create a compact single-image package sufficient for reconstructing a series of images on the receiving end, said Hoffmann’s colleague Voicu Popescu, also a computer science professor and, like Hoffmann, a member of Purdue’s Computer Graphics and Visualization Lab.

More and more, Hoffmann said, scientists want to look at all, or a big part, of a large dataset—data covering genomes, climate change factors, the interactions of elementary particles in physics, the stars in the heavens and more—to glean meaning and ultimately insight in a process he likened to a pyramid.

At the pointy peak, scientists examine large regions of the data at low resolution looking for areas of interest, delve further into them in the middle and discard some false positives while formulating hypotheses and theories, then look at surviving regions of interest at high resolution—the imposing base of the pyramid.

But the pyramid’s base and even the other stages are generally too unwieldy to move long distance via a network, Hoffmann said. So the idea is to cull the data at the source and transmit only the visualization output, images.

Popescu said that’s still no small matter. High-resolution scientific visualizations, and the other kinds of materials the researchers have in mind, can be weighty end products. They’re no YouTube videos. “You can reach very high resolutions,” Popescu said. “The images themselves are non-negligible size, which, multiplied by the frame rate, yields great bandwidth demands.”

Popescu used a movie camera analogy in describing the capabilities of the Purdue system, calling the compressed package “a super frame which would be sufficient to recreate a sequence of frames.” He said the aim is to achieve a highly interactive refresh rate of 60 frames per second or more.

The “camera” in this case is based on a non-pinhole model, he said. Instead of conventionally capturing the data for one frame at a time through a single virtual pinhole, the system records all the data for a visualization segment, that is several frames worth, en masse.

The system takes advantage of the computing power offered by today’s powerful, readily available and largely computer game-driven graphics processing unit chips, or GPUs, to do that with dispatch, and likewise to quickly unpack and display a visualization on the receiving end.

The goal is to make the system responsive enough for interactivity.

Besides visualization, it also might be used for online classes, conferencing and collaboration, allowing a professor, say, at Purdue’s West Lafayette campus to interact more naturally with students at the University’s Calumet campus, in Muncie, Bloomington or elsewhere over something like I-Light, the state’s high-speed, fiber-optic network for linking Indiana’s public and private college campuses, which Purdue and Indiana University jointly manage.

Hoffmann said dealing with video conferencing and collaboration is somewhat trickier because the raw material isn’t digital to start, unlike scientific data and visualizations, and must be converted. The researchers have done some testing of the system across the Purdue campus in West Lafayette. The next step is to try it between Purdue’s West Lafayette and Calumet campuses and over the Northwest Indiana Computational Grid, the scalable, high-speed and high-bandwidth system in Northwest Indiana designed for science and technology research and to enhance economic development in the region. ITaP and the Rosen Center, Purdue Calumet and Notre Dame are partners in the grid.

Calumet created a visualization lab this spring, funded through the Northwest Indiana Computational Grid, with a two-screen immersive virtual environment, haptic equipment to allow users to experience touch and feel when using the facility and supporting features, said Jack Moreland, visualization specialist at the Calumet campus.

Already, researchers at Calumet and steel industry partners have used it to step inside a blast furnace as part of a fluid dynamics simulation exploring ways to improve the manufacturing process, Moreland said. The resource also has been used in teaching chemistry, putting students in the middle of molecules they’re studying.

The system being developed by Hoffmann, Popescu and colleagues could open up numerous other possibilities for research and teaching at Calumet by making it easy to tap data collections and computing power at Purdue’s West Lafayette campus and elsewhere for visualization purposes, Hoffmann and Moreland said.

Writer: Greg Kline, (765) 494-8167, gkline@purdue.edu

Sources: Chris Hoffmann, (765) 494-6185, cmh@cs.purdue.edu
Voicu Popescu, (765) 496-7347, popescu@cs.purdue.edu
Jack Moreland, (219) 989-2765, morelanj@calumet.purdue.edu

Steele supercomputer at Purdue makes list of world's most powerful systems

Tue, 02 Dec 2008

Purdue’s Steele supercomputing cluster—made possible by the collaborative efforts of faculty, staff and volunteers—is among the most powerful high performance computing systems in the world, according to rankings released at the SuperComputing ’08 conference in Austin, Texas, Tuesday (Nov. 18).

The Top 500 Supercomputer Sites project has been ranking the 500 most powerful known computer systems twice a year since 1993 as a way of detecting and tracking trends in high performance computing. Steele placed 105th on the latest list. Purdue ranked 319th in November 2007.

Steele ranked first among the Big 10 universities with systems on the list. Indiana’s Big Red cluster was at 148 and Minnesota had two entries at 268 and 356. The Steele cluster is operated by Purdue’s Rosen Center for Advanced Computing, the research and discovery arm of Information Technology at Purdue, the university’s central information technology organization.

Gerry McCartney, Purdue’s vice president for information technology and chief information officer, said Steele’s showing was important not so much for where it puts Purdue on the Top 500 list as for the trend it indicates.

“Our new supercomputer, Steele, is is another indication that Purdue is once again one of the leaders in high performance computing,” McCartney said. “But of course we don’t do this to see how high we can score on lists such as the Top 500. We do this to enable our scientists and engineers to stay at the forefront of discovery in crucial topics such as cancer, global warming and the lack of affordable energy.”

Purdue is determined to continue enhancing the high performance computing resources it provides for research and economic development purposes across the state, McCartney said. Between 2006 and 2008, ITaP’s Rosen Center increased its computing resources—used by researchers around campus, on Purdue’s satellite campuses and elsewhere—from 14 teraflops, or 14 trillion calculations per second, to 100 teraflops.

A lot of people on the Purdue campus can take some of the credit for Steele’s placement on the list announced at the premier international gathering for high performance computing, networking, storage and analysis. Steele is a “community cluster,” funded by combining faculty grant and lab startup funds and money from institutional sources.

Each “owner” gets a share of the computing power in the machine based on their investment and the opportunity to tap more when they need from the shares of other users idle at the time.

“We built a top 500 machine by working collaboratively with the faculty,” said John Campbell, associate vice president for information technology, who heads the Rosen Center. “This machine is all about pulling together a diverse set of people, utilizing a variety of funding and sharing resources.”

Resources like Steele are integral to the research of Purdue faculty members who helped pay for the cluster, like Gerhard Klimeck, an electrical and computer engineering professor who models the next two or three generations of nanoscale electronic devices, allowing their properties to be understood long before they’re ever fabricated.

Atoms, electrons—the things Klimeck looks at are very small. But there are millions, maybe billions, of them moving and interacting in myriad ways. That makes simulating them as complicated as simulating something very big, the cosmos for instance.

“We’re compute driven,” Klimeck said.

Moreover, building Steele was a community effort, too. More than 250 staff members and volunteers assembled the cluster in a single Monday morning in May. Some of them even came from Purdue’s diehard in-state athletic rival Indiana, attracted by the idea of a high-tech barn raising to assemble an 18 wheeler-sized supercomputer in a day, a process that normally takes weeks.

Campbell noted that Steele recently averaged 87 percent owner utilization and more than 98 percent utilization overall.

That’s one reason the Rosen Center already is planning Purdue’s next community cluster, to be built in the spring of 2009. Faculty and campus organizations interested in participating in the new cluster, to be called Coates, can find more information at: http://www.rcac.purdue.edu/userinfo/resources/coates/.

The Top 500 list generally captures the most powerful academic and commercial research supercomputers. Information about classified systems, used largely for defense purposes, isn’t released for the list.

Steele, made from 893 Dell 1950 systems, also prompted the appearance of a Purdue logo on a slide highlighting key Dell customers in the SC08 conference keynote talk Tuesday morning by Michael Dell, chairman of the board and chief executive officer of the world's second largest computer company.

As with Steele, the Rosen Center is naming its new cluster after an important figure in the history of computing at Purdue. Clarence L. “Ben” Coates headed the School of Electrical Engineering (now Electrical and Computer Engineering) and was a driving force behind creating a high performance computing network for Purdue’s engineering schools. John Steele was instrumental in founding and served as the director of the Purdue University Computing Center, the high performance computing unit at Purdue prior to the Rosen Center.

The testing of Steele for supercomputing’s Top 500 list required the running of the High-Performance LINPACK benchmarking suite, software designed to stress three big computationally oriented components of a cluster—its processing power, memory and the wiring that links its processors together for working in concert.

The Rosen Center’s Andy Howard, who volunteered for the benchmarking job, said it is possible simply to download the code and run it on most machines, “but it’s not going to be your best score.”

Like a lot of software, many scientific applications included, the benchmarking suite runs best with some relatively minor tuning for conditions on a specific machine, the number of processors for instance. Howard, a senior in electrical and computer engineering technology from West Lafayette who works as an assistant research programmer for the Rosen Center, tweaked the program in six test sessions spread over three weeks.

Steele is strong from the perspective of its processing power and memory, Howard said. Where it comes up a little short is in the connective wiring, which doesn’t include a lot of expensive low-latency hardware because that isn’t a feature much in demand by the researchers who use, and paid for, the system.

Writer: Greg Kline, (765) 494-8167, gkline@purdue.edu

Sources: Gerry McCartney, (765) 496-2270, mccart@purdue.edu
John Campbell, (765) 494-1289, john-campbell@purdue.edu

Purdue project aims to put isotope analysis on the map, and the web

Tue, 02 Dec 2008

Mapping and spatial analysis of isotopes in water, particularly hydrogen and oxygen isotopes, could be used to trace the ultimate source of a city’s water supply, the wintering sites of migrating birds, the trading patterns of prehistoric peoples, perhaps even the travels of an unidentified corpse to a crime scene.

Researchers at Purdue, its Earth and Atmospheric Sciences, Library and Statistics departments and the University’s Rosen Center for Advanced Computing are aiming to make it easer for ecologists, environmental scientists, hydrologists and many others, including policy-makers, to use the isotope data for a variety of purposes.

They’re working on a cyberinfrastructure for tapping an extensive, distributed data collection on line, and related Web-based tools allowing real science to be done with it, under a three-year, $831,573 grant from the National Science Foundation, its Division of Biological Infrastructure and the National Ecological Observatory Network.

The system focused on modeling stable isotope data with respect to space and time also will take advantage of the computational and storage resources available through the TeraGrid, the world's largest open science computing network, which is funded by the NSF as well. Purdue through the Rosen Center is a TeraGrid partner and resource provider.

“You need to make it easy for people to use (the isotope data) and incorporate it in their own research,” said Purdue earth and atmospheric sciences Professor Gabriel Bowen, who employs such data in his own research on large-scale, regional to global environmental change. He is a member of the Purdue Climate Change Research Center and co-directs the Purdue Stable Isotope Facility.

The project will include online tutorials to facilitate data access and use by scientists and non-scientists alike.

Water being the stuff of life, and hydrogen and oxygen the stuff of water, that makes hydrogen and oxygen isotopes a handy, useful and near ubiquitous avenue for the research Bowen conducts, which includes potential forensic uses as well as environmental science, and a wide range of other science and policy applications.

The mix of isotopes—atoms with variances in their number of neutrons—varies according to where those atoms have been, allowing someone like Bowen, for example, to trace the water coming from a tap in a town to snow melt from a mountain peak by reading, in effect, its isotope fingerprint. Likewise, water isotope tracers can be used in similar fashion to determine the geospatial origins of biological and geological materials.

In a sense, INPort, it stands for Isotopes Network Portal, will link a large range of scientists and non-scientists transparently to a vast fingerprint file of isotope data with integrated querying and geospatial modeling operations akin to a Geographic Information System, or GIS. Among other things, it will be able to produce maps showing the distribution of isotopes geographically and over time.

Besides Bowen and Lan Zhao, a Rosen Center research scientist, principal investigators on the project include statistics Professor Tonglin Zhang, library science Professor Christopher Miller, a GIS specialist, and Jason West, an ecosystem science and management professor at Texas A&M University, with whom Bowen worked when the two were postdocs.

Bowen and colleagues, in addition to drawing on 40 years of rainwater and snowfall measurements collected at several hundred sites worldwide by the International Atomic Energy Agency and the World Meteorological Organization, already have put together a large database of regional isotope observations and spatial statistical analysis tools people are using for teaching, research and regulatory purposes.

When he began looking at expanding on the effort and making the data more accessible and easier to use, someone recommended Zhao and the Rosen Center’s scientific solutions group led by Carol Song, which has done similar science “gateway” projects, for instance a Web-based climate modeling system for TeraGrid users. Their approved proposal started in August 2008. Bowen said the plan is for an initial deployment of INPort within the next year.

Zhao said the intent also is to leverage her group's work in serving earth science data to researchers by creating software components that can be used with existing systems and for future projects with similar requirements in biosciences, geosciences and other fields.

The Rosen Center is the research and discovery computing arm of Information Technology at Purdue (ITaP), the University’s central information technology organization.

Writer: Greg Kline, (765) 494-8167, gkline@purdue.edu

Sources: Gabriel Bowen, (765) 496-9344, gabe@purdue.edu
Lan Zhao, (765) 496-2079, lanzhao@purdue.edu

More information: waterisotopes.org www.rcac.purdue.edu

Purdue team learns from Bandwidth challenge at supercomputing conference

Tue, 02 Dec 2008

A team of staff members from Purdue’s Rosen Center for Advanced Computing learned some things likely to come in handy from competing in the Bandwidth Challenge at SuperComputing ’08, the world’s premier high performance computing conference.

Purdue was one of six finalists among an international collection of entries. University of Illinois-Chicago won the competition at the conference, known as SC08, held in Austin, Texas, this week. The placement of the other finalists wasn’t revealed by the event’s organizers when the winner was announced Thursday (Nov. 20).

“The Bandwidth Challenge is an excellent opportunity for Purdue to experiment with wide-area storage for later use in storage traffic between our regional campuses and our West Lafayette central campus,” said Michael Shuey, high performance computing technical architect at the Rosen Center.

Besides being useful in sharing resources with Purdue’s satellite campuses, the technology tested by the team also might be applicable in an off-campus data center if Purdue builds one in the future.

Purdue's Bandwidth Challenge team included Shuey; Ramon Williamson, the senior storage engineer at the Rosen Center, who served as team manager; Greg Veldman, storage administrator, and Patrick Finnegan, UNIX system administrator. The team partnered with BlueArc, whose server computers the Rosen Center uses to manage data traffic, Texas Memory Systems, makers of fast solid-state storage units akin to a USB flash drive, and Foundry Networks, which provided the high-speed network connection to the SC08 network in Austin, and hence to the outside world, for the competition.

The Rosen Center is the research and discovery arm of Information Technology at Purdue, the University’s central information technology organization.

The Bandwidth Challenge pushes the limits of moving data over a long-distance computer network, such as the Internet or the TeraGrid, the world's largest network for open science computing. Purdue is a partner in and resource provider on the TeraGrid, which is funded by the National Science Foundation.

Participants in the challenge were judged on the peak amount of data they moved from one place to another during the competition and how high a sustained rate they maintained, as well as their approach to “real-world applications and data movement issues.” Purdue used free software and commodity hardware in the effort, working in part with simulated climate data, which the team moved between West Lafayette and Austin.

Contact: Greg Kline, (765) 494-8167, gkline@purdue.edu

Sources: Ramon Williamson, (765) 496-1272, ramonw@purdue.edu
Michael Shuey, (765) 494-0985, shuey@purdue.edu

Purdue cluster challenge team creates a stir at supercomputing conference

Tue, 02 Dec 2008

Purdue’s team may not have won the Cluster Challenge at the SuperComputing ’08 Conference in Austin, Texas.

Then again, no other team got to run code on a supercomputer powered by bicycling colleagues of seven-time Tour de France champion Lance Armstrong.

The Purdue team also was the first to break the teraflop barrier—a trillion calculations per second—in the student event at the premier international high performance computing conference.

Cyclists from Mellow Johnny’s Bike Shop in Austin, of which Armstrong is part owner, pedal powered a low-power “green” supercomputer from Massachusetts-based SiCortex, Purdue’s partner in the Cluster Challenge, on Thursday as it ran code prepared by the Purdue team.

The program was a retooled version of an historic mathematical simulation of a dam under stress, a problem first solved 100 years ago by a team of human “computers” who worked on it for three years using slide rules. The SiCortex took about 10 minutes in a demonstration that drew a large crowd.

The Purdue team sponsored by Information Technology at Purdue, the University’s central information technology organization, and the Rosen Center for Advanced Computing, ITaP’s research and discovery arm, used a SiCortex for the Cluster Challenge.

Team member Andy Howard, a senior in electrical and computer engineering technology from West Lafayette, said he and his teammates were able to use the competition compliment of about 1,000 processors in the machine to run most of their jobs simultaneously the bulk of the time. But the simulated data, unlike that used in last year’s Cluster Challenge, wasn’t as suited to a processor-rich machine like the SiCortex.

The Purdue team did finish first in the benchmarking phase of the competition Monday, designed to test the limits of the entries at the outset, ahead of a combined team from Dresden University in Germany and Indiana University, which won the overall competition.

The teams are judged on their benchmarking score, the amount of data they can process with designated applications over three days and their overall presentation, preparation and knowledge.

They have to run on a strict 26-amp power limit, too. But shortly after the competition, with the power limit off, the Purdue team and SiCortex beefed up their machine to its full compliment of more than 1,400 processors and ran the benchmarking software again, hitting slightly over a teraflop.

SiCortex builds supercomputers designed to deliver high performance using large numbers of slower, energy-efficient processors, both in power consumption and the cooling they require. Among other things, the company uses a unique, very fast “interconnect fabric”—the wiring that links its processors for working in concert—offsetting the raw speed disadvantage.

Purdue's team consists of Howard; Alex Younts, a sophomore in computer science from West Lafayette; David King, a senior in electrical and computer engineering technology from Lafayette; Paul Willmann, a senior in computer technology from Carmel; and Ryan Weinschenk, a senior in electrical and computer engineering technology from Noblesville.

The students are enrolled in a high performance computing class taught by Jeffrey Evans, an assistant professor in the Department of Electrical and Computer Engineering Technology. Preston Smith, senior UNIX system administrator for Rosen Center, served as team leader.

For the major portion of the competition, Purdue’s and the other Cluster Challenge teams had to run a battery of scientific applications used for such purposes as examining the seismic waves from earthquakes, pollution from automobiles, air flow over a jet plane, evolution at a genetic level and more.

The applications were announced earlier this year, giving the teams months to work with them in the context of their competition machines. In Purdue’s case the unusual architecture of the SiCortex required some adjustments in the software, especially to take maximum advantage of the computer’s voluminous processor array.

The teams didn’t see the simulated data to be run in the applications until it was downloaded to their machines minutes before the biggest part of the competition began Monday evening. That kicked off what Ricky Kendell, Indiana native and Cluster Challenge co-chairman, called “a 44-hour tenacity test.”

“We give them enough work to take seven or eight days and they get 44 hours to do that,” Kendell said.

Smith said the amount of data is, in fact, generally too much for any team to finish. The goal is to get as far as possible. The Purdue team, working in shifts, kept at that task until late Wednesday afternoon when the competition ended.

“I slept part of the time,” said King, coming off the overnight shift Tuesday, although he allowed that the rock music blaring near constantly from loudspeakers next to the Cluster Challenge area at the conference made sleeping a challenge as well.

“I went to the show floor and I could still hear it in my head,” Younts said.

The students normally didn’t have a lot to do once they started a data processing job.

“If something goes wrong we have to fix it,” Weinschenk said. “But until something goes wrong, it’s mainly just sitting here waiting for something to go wrong. It’s kind of like pulling an all-nighter, but not actually doing anything.”

Still, tired or not, they were ready to head to a party thrown for the teams at the end of the competition.

“Eight hours of sleep in the last 48 hours, that’s not bad,” Howard said.

Writer: Greg Kline, (765) 494-8167, gkline@purdue.edu

Sources: John Campbell, associate vice president for information technology, (765) 494-1289, john-campbell@purdue.edu
Jeffrey Evans, (765) 494-7725, jje@purdue.edu
Preston Smith, (765) 494-9729, psmith@purdue.edu
Andy Howard, (765) 496-2434, ahoward@purdue.edu
Alex Younts, (765) 496-2023, ay@purdue.edu
David King, (765) 491-8158, dk@purdue.edu
Ryan Weinschenk, electrical and computer engineering technology, (317) 690-1906, cweinsch@purdue.edu
Paul Willmann, computer technology, (317) 496-2067, willmann.paul@gmail.com

Purdue Cluster challenge team starts fast

Tue, 02 Dec 2008

Purdue’s Cluster Challenge team started off fast at the SuperComputing ’08 Conference in Austin, Texas, Monday evening (Nov. 17).

As in 694 gigaflops fast.

That’s the score the Purdue team posted with the unusual SiCortex supercomputer it is using in the competition. The score was posted running the HPC LINPACK benchmarking software, the first step for all the teams in the Cluster Challenge.

Purdue placed first in the benchmarking segment overall and had the second highest LINPACK score, behind only National Tsing Hua University of Taiwan, which posted a score of 703 gigaflops. A combined team from Indiana University and Dresden University in Germany was third with a score of 526 gigaflops. None of the other four teams topped 500.

Purdue's team consists of Andy Howard, a senior in electrical and computer engineering technology from West Lafayette; Alex Younts, a sophomore in computer science from West Lafayette; David King, a senior in electrical and computer engineering technology from Lafayette; Paul Willmann, a senior in computer technology from Carmel; and Ryan Weinschenk, a senior in electrical and computer engineering technology from Noblesville.

The students are enrolled in a high performance computing class taught by Jeffrey Evans, an assistant professor in the Department of Electrical and Computer Engineering Technology. Preston Smith, senior UNIX system administrator for Purdue's Rosen Center for Advanced Computing, is serving as team leader.

The team, working in shifts, is now in a 44-hour marathon running simulated data in a battery of scientific applications relevant to fields ranging from genetics to designing jet airplanes. The competition ends Wednesday.

The teams are judged on their benchmarking score, the amount of data they can process with applications in the time allotted and their overall presentation, preparation and knowledge.

SiCortex builds supercomputers designed to deliver high performance using large numbers of slower, energy-efficient processors, both in power consumption and the cooling they require. Among other things, the company uses a unique, very fast “interconnect fabric”—the wiring that links its processors for working in concert—offsetting the raw speed disadvantage.

The Rosen Center, the research and discovery arm of Information Technology at Purdue (ITaP), the University’s central information technology organization, installed a top-of-the line SiCortex machine in June to test it from both performance and energy-saving perspectives, making Purdue the first university to do so. The SC1458 the Cluster Challenge team is using is the next step down from the Rosen Center’s SC5832. The numbers in both cases refer to the maximum number of processors the machines can contain.

Writer: Greg Kline, (765) 494-8167, gkline@purdue.edu

Purdue and IU become computing teammates

Tue, 02 Dec 2008

Purdue University is adding to what is already the world's largest science-focused distributed computing system with the addition of computers from athletics rival, but longtime computing and research collaborator, Indiana University.

IU is linking nearly 5,000 computers from its research pool to Purdue's national computing grid, which is known as Diagrid. Notre Dame, Indiana State University and Purdue's Calumet regional campus are already contributors in the effort in the Hoosier state.

<> For complete story see: http://news.uns.purdue.edu/x/2008b/081118McCartneyPool.html

Visitors at supercomputing conference impressed with Purdue-developed technology

Tue, 02 Dec 2008

A Purdue-developed technology called HUBzero sounds chilly, but it looks to Chris McPhee like a hot way to allow researchers to do science on line easily in a graphical Web-based fashion, rather than typing in cryptic commands DOS style.

“We’re tired of people seeing black screens,” said McPhee, a system engineer from Queen’s University in Ontario, Canada, who stopped by the Purdue booth at the SuperComputing ’08 conference in Austin, Texas, for a presentation on HUBzero.

The presentation by Michael McLennan, senior research scientist and hub technology architect at Purdue, caught McPhee’s eye as he walked by the Purdue booth Tuesday at the conference, known as SC08.

“I thought, ‘That’s something we might be interested in,’” said McPhee, representing a consortium of Canadian universities at the premier international gathering of high performance computing experts, which runs through Friday.

Presentations by Purdue researchers at the booth and in other conference events have covered a variety of topics on the cutting edges of high performance computing, from an easy way to do advanced science in a Web-based environment today to getting ready to take advantage of tomorrow’s next-generation high performance computers.

The booth at SC08 is designed to promote Purdue, Information Technology at Purdue (ITaP), the University’s central information technology organization, and the Rosen Center for Advanced Computing, ITaP’s research and discovery arm. The conference has more than 10,000 attendees this year.

The theme of the Purdue booth is: “No cycle left behind, no byte left unexplored,” highlighting creative ways Purdue, ITaP and the Rosen Center are finding to improve scientific productivity, said John Campbell, associate vice president for information technology, who heads the Rosen Center.

For example, Purdue’s more than 20,000-processor strong Condor pool, a high-throughput distributed computing system that makes use of otherwise idle machines in offices, labs and elsewhere for research purposes, not only on campus but internationally through the Open Science Grid and the TeraGrid, the National Science Foundation-funded world’s largest network for open science computing.

Purdue this week announced that it is adding to what’s already the world's largest science-focused distributed computing system with the addition of computers from athletics rival, but longtime computing and research collaborator, Indiana University.

IU is linking nearly 5,000 computers from its research pool to Purdue's national computing grid, which is known as Diagrid. Notre Dame, Indiana State University and Purdue's Calumet regional campus are already contributors in the effort in the Hoosier state. Diagrid also will soon include computers from Indiana University-Purdue University Fort Wayne and Purdue North Central.

Like HUBzero, Purdue’s use of Condor, the system underlying Diagrid, also attracted a lot of interest at SC08. Central Florida University just installed its own high performance computing center, but like Purdue and other schools a lot of compute cycles elsewhere around campus go to waste, said Ravi Palaniappan and Joe Sottilare of Central Florida, which may look to Purdue’s model to capture a good number of those.

“One of the things we were told to look at was using Condor to harvest cycles,” said Palaniappan, who added that he was “given explicit instructions” to check out the SC08 talk on using Condor by Preston Smith, the Rosen Center’s senior UNIX administrator.

HUBzero was developed for nanoHUB, an online community and toolbox for the nanotechnology development and education. But Purdue realized the underlying technology could be extracted for use by all sorts of scientific, technical and research communities, said Gerry McCartney, Purdue’s vice president for information technology and chief information officer. McCartney likened HUBzero to a great design for a library, just waiting for books to fill the shelves.

In HUBzero’s case, those books are scientific simulation and modeling tools in particular, but also things like interactive online classes and tutorials, question and answer forums, podcasts and more, McLennan said. More than a half dozen hubs, in fields ranging from cancer care and accelerating health care research in general to advanced manufacturing techniques and global engineering education, have now been developed using the technology.

In a packed presentation on nanoHUB at Purdue’s SC08 booth, Purdue electrical and computer engineering Professor Gerhard Klimeck said the resource now has 85,000 users in 172 countries, and growing. The more than 120 simulation tools on nanoHUB, which users can employ immediately without installing any software, are being used for real science, Klimeck said. He noted that nanoHUB simulations have been cited more than 256 times in scientific literature, mostly peer-reviewed journals and proceedings. Klimeck also is the technical Director of the Network for Computational Nanotechnology at Purdue.

More information:
http://hubzero.org/
http://www.nanohub.org/

Writer: Greg Kline, (765) 494-8167, gkline@purdue.edu

Sources: Gerry McCartney, (765) 496-2270, mccart@purdue.edu
John Campbell, (765) 494-1289, john-campbell@purdue.edu
Michael McLennan, (765) 494-6495, mmclennan@purdue.edu
Gerhard Klimeck, (765) 494-9212, gekco@purdue.edu

Virtual clean room to enhance training for pharmacy students

Tue, 02 Dec 2008

Editor's note: View a fly-through animation of the clean room from the Rosen Center Web site, http://www.rcac.purdue.edu/news/video/cleanroom.cfm.

Hard-to-come-by training time in a pharmacy clean room is about to become a lot easier for Purdue pharmacy students to get, eventually as easy as turning on their laptop computers.

Pharmacy clean rooms are sterile environments where pharmacists and pharmacy technicians prepare materials that need to be guaranteed contamination free, said Steve Abel, assistant dean for clinical programs in the Purdue School of Pharmacy and Pharmaceutical Sciences.

Researchers from Purdue’s Envision Center for Data Perceptualization, part of the Rosen Center for Advanced Computing and Information Technology at Purdue (ITaP), the School of Pharmacy and Pharmaceutical Sciences, and the School of Industrial Engineering are creating a virtual version of a standard hospital clean room.

The prototype runs in a multiwall immersive environment at the Envision Center and will work on wall-sized panels and portable systems as well. The equipment employs 3-D glasses and wireless controllers to put users in the middle of the virtual world being projected and allow them to navigate and manipulate it.

Meanwhile, undergraduate research assistants for the Envision Center also are adding the capability for modifying the virtual pharmacy clean room to run on Unreal Engine 2, software underlying the popular computer game, itself a stroll—or panicked sprint—through a virtual environment, albeit with opponents shooting at you. The additions will allow the virtual clean room to be used on a PC.

The game engine is openly available for professional developers and hobbyists to build other worlds on top of, largely for games, although its first-person perspective is ideal for something like the virtual clean room as well. For example, researchers elsewhere have used it to create a virtual control room for training workers at nuclear power plants.

Generally found in hospitals and home health care companies, clean rooms are used to prepare drugs, intravenous drips, syringes, chemotherapy treatments and the like, especially those administered through the veins, a factor that makes use of a clean room and proper clean room procedures vital.

“You’re putting the medication right in (patients’) blood so it goes right to their organs,” Abel said.

Concern over the rise of antibiotic-resistant pathogens has only increased the need for expertise in the use of clean rooms.

“It’s really important that pharmacists be trained to use them,” Abel said.

But the number of facilities where pharmacy students can train is limited. When the training involves real materials, it also can be expensive, sometimes prohibitively so.

Abel said Purdue pharmacy students tend to get the training at the end of their third year, just before their last year, in which they serve a professional practicum.

“A lot of it is classroom,” he said of the clean room training. “Just a small amount of it is actual hands on.”

Moreover, students often don’t have the easiest time getting used to the clean room environment once they’re in it. They’re covered in gowns, shoe covers, caps, and gloves. There are needles involved. The work of mixing and packaging materials is exacting. That the situation can be nerve-racking at first isn’t surprising.

Abel said the plan is to have students work in the virtual clean room clad as they would be in a real-life clean room. Unlike a real clean room, the system will warn them when they make a mistake, allowing them to do a procedure over without the potential for tragic consequences.

The virtual clean room is designed to give pharmacy students plenty of “stick time”—in the vernacular of pilots, who do a lot of virtual flight training on computerized simulators these days—even before they set foot in a real clean room.

The Envision Center uses advanced data visualization and perceptualization tools to process and display scientific information in ways that make complex phenomena easier to comprehend and enable the discovery of new knowledge and the development of innovative products.

Researchers at the center explore novel computer graphics and human-computer interface technologies and integrate them with state-of-the-art advanced computation and networking and high-end immersive environments.

Envision Center Managing Director Steve Dunlop and Abel connected in an Envision Center tour for a campus guest Abel hosted. He and Dunlop ended up brainstorming about what the center’s work with virtual environments might do for pharmacy teaching, learning, and training.

The idea for a virtual clean room came out of the discussions between Abel and Dunlop and they, along with industrial engineering Professor Leyla Ozsen, obtained a Purdue Provost’s instructional grant, as well as some funding from Eli Lilly and Company and Purdue’s Pharmacy School, to pay for the development.

From there, Dunlop went looking for students to work on the project. The attraction: they’d get paid, but more importantly have something impressive for their portfolios that people would use. The undergraduate research assistants working with Dunlop include Chris Mankey of Fishers, Chris Sprunger of Lafayette, and Evan Underwood of Kokomo, all from the Computer Graphics Technology Department.

“This is a great opportunity, to actually work on something that’s relevant and meaningful,” Mankey said.

The team started work in February, among other things visiting multiple clean rooms at Clarion Health Partners and Wishard Health Services in Indianapolis.

“We took digital pictures from multiple angles to provide the necessary details,” Underwood said.

Those pictures became models for creating the elements of the virtual environment, using high-end computer graphics programs like Maya and 3ds Max, right down to individual vials of materials.

Underwood said they started with “primitive shapes,” your basic cube for instance, morphed them to the shape of the object they were creating and added relevant coloring and texture. They also added shading and lighting, based on the lighting of the hospital facilities they visited.

The object creation was the most laborious part of the process, the students said, although it helped that once they had one, say, IV bag created it could be copied and repurposed elsewhere in the virtual clean room.

The simulation includes motion-tracking capability that adjusts the view as users move their heads or “walk” through the rooms—a clean room facility is actually more than one room—using the wireless controller.

“You can change the perspective to create a more lifelike environment.,” Sprunger said.

Laura Arns, associate director of the Envision Center, is providing technical expertise on the development of the interaction in the virtual environment, the picking up, moving, and manipulation of objects. While not in the first version, the Envision Center might eventually add haptics capability to the simulation, that is touch and feel, which could let a user experience the weight and texture of something picked up in the virtual environment.

In addition, Ozsen and her students are to use to the computer model to study clean room design and work flow to see where improvements might be made.

The computer graphics students also created a movie “fly-through” of the virtual clean room that can be used for presentations about the project and appeared at the Purdue booth at SIGGRAPH, the major annual computer graphics conference, held Aug. 11-15 in Los Angeles this year.

The movie, 1 minute 58 seconds long, incorporates 3,060 frames, any one of which takes about an hour to render on a single computer.

The students cut the rendering time down to about 48 hours, however, by using TeraDRE. Arns and David Braun of the Rosen Center provided assistance with using the distributed rendering environment, which draws on the Purdue Condor pool, a system for harnessing unused compute time on nearly 20,000 processors on campus, along with computers at Purdue Calumet, Indiana, Indiana State and Notre Dame. It works over the TeraGrid, the world’s largest open computing network for scientific research, in which Purdue, via the Rosen Center, is a partner.

Writer: Greg Kline, (765) 494-8167, kline@purdue.edu

Sources: Steve Abel, (317) 613-2315, sabel@iupui.edu
Steve Dunlop, (765) 494-5861, dunlops@purdue.edu
Laura Arns, (765) 496-7888, arns@purdue.edu

Making sense of the world an old idea, new technologies offer ways to do it better than ever

Tue, 02 Dec 2008

A Babylonian clay tablet dating from 600 B.C. is the oldest map of the known world, although not a whole lot of the world was actually “known” at that point.

The Chinese, Egyptians and Mesopotamians used string and bead abacuses to make calculations at least as early as 3000 B.C. Ancient Alexandria boasted data collections—in the form of a library filled with papyrus scrolls.

When the brothers Montgolfier first got a look at the ground in France from the vantage of their hot air balloon in 1783, they were engaging in remote sensing, in a sense. In 1903, German engineer Julius Neubronner combined a small camera with a mechanical timer strung around a homing pigeon's neck for aerial photography. Before long, airplanes were carrying the cameras.

Today, “spatial data infrastructures” national, regional and global in scale rely on a satellite’s sensor suite view of the world rather than a bird’s-eye view. The systems employ and integrate myriad data collections from satellite and other remote sensing techniques—as well scientific and demographic material, for example, collected on the ground—sifted by powerful supercomputers.

But the idea is the same: to make some sense of the world. And no longer just for the U.S. and fellow developed nations, which have long experience building and using such tools for military, scientific and other purposes, said Gilbert Rochon, who directs the Purdue Terrestrial Observatory, part of Information Technology at Purdue (ITaP) and its Rosen Center for Advanced Computing.

Spatial data infrastructures are appearing in, and have relevance for, places such as Argentina and Brazil, Nigeria and Thailand, in addition to less surprising venues like tech-savvy China, India and Israel. More than a dozen countries now have middle- to high-resolution Earth-observing satellites in orbit and the number should top 20 by decade’s end. China, Japan, Australia and other Pacific Rim nations are working on a regional system as is the European Union.

“There’s a big push on for a global spatial data infrastructure,” said Rochon, who also is associate vice president for collaborative research in ITaP and chief scientist for the Rosen Center.

Rochon has been discussing the future of spatial data infrastructure technology and its implications worldwide recently.

In October, he was a keynote speaker at the Korean national conference on spatial data infrastructures. Purdue agronomy, civil engineering and electrical and computer engineering Professor Melba Crawford, assistant dean of engineering for interdisciplinary research and director of the Laboratory for Applications of Remote Sensing at the University, also presented a paper at the conference, on advances in LIDAR (for light detection and ranging), an optical remote sensing technology that measures properties of scattered light to generate information about a target.

Rochon in October also spoke about a potential use of spatial data infrastructures—early warning and mitigation of vector-borne and zoonotic diseases—at the African Association for Remote Sensing of the Environment meeting in Accra, Ghana. The first author of the paper, Joseph Quansah, earned his doctorate at Purdue in agricultural and biological engineering and is a postdoctoral researcher here.

You can think of a spatial data infrastructure as a geographic information system writ large, or “enterprise GIS,” as Rochon put it. More than just maps, the idea is a unified system for ready access to standardized data from a variety of sources that can be rendered visually and geographically to generate knowledge and aid decision making at all scales for multiple purposes.

“There are a host of potential applications,” Rochon said.

Those applications range from routine local governance issues, like planning the location of a new fire station and other land use issues, to illuminating the effects of global climate change. Changes in land use like deforestation and urban sprawl, the impact of weather and natural disasters, socio-economic and demographic trends, pollution and more—all could be brought into the perspective by the techniques and, perhaps, be better managed as a result.

But Rochon said the grand challenge for the technology is really to apply it to addressing issues important to humanity, the spread of disease or famine, food and water security, sustainable development, poverty, disaster mitigation, and mitigating the effects of armed conflict among them.

Inclusion and equitability for developing nations and rural areas, which may lack the necessary financial and human resources and essential components like fiber-optic networks, high performance computing facilities and satellite tracking stations, even an adequate power system, also are a challenge, Rochon said.

Purdue has been involved in remote sensing research for more than 40 years. The Purdue Terrestrial Observatory gathers masses of satellite and other remote sensing data along with information collected from ground-truthing instruments and activities, which allows the data to be mined for knowledge by researchers from an interdisciplinary array of fields as well as by local, state and national decision makers.

Resources like the Purdue Real-time Satellite Information Gateway (PRESTIGE) and Purdue-developed HUBzero (an easy-to-use, yet sophisticated Web portal development system) can be used to make the data and tools to use it readily available via the Web and the TeraGrid, the world’s largest network for open science computing.

Rochon noted that Purdue is the lead institution in IndianaView, a statewide consortium of universities, state organizations and non-profits, which is a member of AmericaView, a nationwide program that focuses on satellite remote sensing data and technologies in support of applied research, education, workforce development and technology transfer for what is essentially a U.S. spatial data infrastructure.

More information:
www.itap.purdue.edu/pto
www.lars.purdue.edu
www.purdue.teragrid.org/prestige
hubzero.org

Writer: Greg Kline, (765) 494-8167, gkline@purdue.edu
Source: Gilbert Rochon, (765) 496-2274, rochon@purdue.edu

Pilot program lets users try Sun systems before making purchases

Tue, 02 Dec 2008

ITaP and Purchasing announce a pilot program designed to give researchers at Purdue rapid access to new computational hardware.

The program also allows faculty and staff to test a variety of systems before making purchases. Since the equipment prices have been pre-negotiated, the program can reduce researchers’ “time to science” on purchases from weeks to days.

Sun Microsystems will maintain a small inventory at Purdue that can be made available to purchasers almost immediately. Larger orders should typically take just a few days for delivery. The equipment can be supported by Purdue’s Rosen Center for Advanced Computing as a build-to-order cluster or shipped to a researcher’s specified location.

The initial program offers high performance Sun systems configured for both compute- and data-intensive needs. These systems are available for testing and purchase through the Rosen Center. Faculty and staff members who decide to purchase any of these models will be able to do so at a reduced price under the program involving Purdue, Sun reseller Dewpoint and Sun.

The Rosen Center has set up a four-node cluster of Sun Fire X2220s that’s available for testing.

Other machines can be set up quickly by request. Researchers who test a machine are under no purchase obligation, nor do they have to test to buy.

For information on the models available through the Rosen Center visit: www.rcac.purdue.edu/hardwarepurchase/sun.cfm.

For more information, email rcac-help@purdue.edu.

The Rosen Center for Advanced Computing (RCAC) is the research and discovery arm of Information Technology at Purdue (ITaP), the University's central information technology organization. RCAC supplies reliable and secure high performance computing systems and storage to faculty and staff. This allows researchers doing computationally intensive cutting-edge science, engineering and social science research to concentrate on their work rather than maintaining a computer system.

Purdue Cluster Challenge team aiming to take technology to limits

Tue, 02 Dec 2008

Two groups of Purdue University students and staff will compete in events aimed at pushing the limits of technology to meet the needs of the scientific community.

A team of five students enrolled in a high performance computing class at Purdue is gearing up for the Cluster Challenge, scheduled for Nov. 17-19 at the SuperComputing '08 Conference in Austin, Texas. Another Purdue team of four professionals from Purdue's Rosen Center for Advanced Computing will compete in an event called the Bandwidth Challenge.

The premier international conference for high performance computing, networking, storage and analysis runs Nov. 15-21, and 11 universities will compete in the challenges during the event.

"Our challenge entries are all about pushing the technology to meet the demands of the scientific community," said John Campbell, associate vice president for information technology, who heads the Rosen Center.

Cluster Challenge teams must run a battery of benchmarking programs and working applications on their machines as efficiently as possible. Purdue's team has spent most of the semester preparing to run a wide range of scientific software within the constraints of the contest. Rules limit the amount of power the entries can draw from two 120-volt, 13-amp circuits, which limits the number of processors that can be employed.

Purdue's competition machine has been in Boston, where partner SiCortex is located, and the group has been working remotely - as if the computer was in the same room. Most of the team's work has involved retooling software to run on its entry in a way that takes maximum advantage of its voluminous array of processors.

SiCortex builds unusual supercomputers designed to deliver high performance by using thousands of slower, energy-efficient processors, both in power consumption and cooling. Among other things, the company uses a unique, very fast "interconnect fabric" - the wiring that links its processors for working in concert - offsetting the raw speed disadvantage.

"It's not your standard architecture, so you can't just install it and it works," said team member Ryan Weinschenk, a senior in electrical and computer engineering technology from Noblesville. "The hardware performance is there. It's just actually getting the software to run."

Purdue's team consists of Andy Howard, a senior in electrical and computer engineering technology from West Lafayette; Alex Younts, a sophomore in computer science from West Lafayette; David King, a senior in electrical and computer engineering technology from Lafayette; Paul Willmann, a senior in computer technology from Carmel, Ind.; and Weinschenk.

The Students are enrolled in a class taught by Jeffrey Evans, an assistant professor in the Department of Electrical and Computer Engineering Technology.

The Bandwidth Challenge pushes the limits of moving data over a long-distance computer network, such as the Internet or the TeraGrid, the world's largest network for open science computing. Purdue is a partner in and resource provider on the TeraGrid, which is funded by the National Science Foundation.

Participants in the challenge are judged on the peak amount of data they can move from one place to another and how high a sustained rate they can maintain. This year, the competition also is placing a premium on "real-world applications and data movement issues" in addition to merely filling the pipeline; however. Purdue will try to meet the challenge using free software and commodity hardware.

Purdue's Bandwidth Challenge team includes Ramon Williamson, the senior storage engineer at the Rosen Center for Advanced Computing, who is serving as team manager; Michael Shuey, high-performance computing technical architect; Greg Veldman, storage administrator, and Patrick Finnegan, UNIX system administrator. The team is partnering with BlueArc, Texas Memory Systems, makers of fast solid-state storage units akin to a USB flash drive, and Foundry Networks, which is providing the high-speed network connections between the storage appliances and the SC08 network in Austin for the competition.

What the Rosen Center is learning from the exercise, besides being of value to other high-performance computing centers, should prove useful at home, Williamson said.

A cluster to be built at Purdue Calumet from parts of West Lafayette's decommissioned Lear supercomputer is likely to be accessing storage on the West Lafayette campus and could employ something like the system developed by the Bandwidth Challenge team.

In addition, at some point the Rosen Center, pressed for space in its machine room in the Mathematical Sciences Building, might create a central, remote storage facility in the West Lafayette area or elsewhere, which could function, using the techniques, as if it were on site, Williamson said.

Writer: Greg Kline, (765) (765) 494-8167, gkline@purdue.edu

Sources: John Campbell, (765) 494-1289, john-campbell@purdue.edu

Jeffrey Evans, (765) 494-8167, jje@purdue.edu

Ryan Weinschenk, (317) 690-1906, cweinsch@purdue.edu

Purdue News Service: (765) 494-2096; purduenews@purdue.edu

Game lets geeks compete to build virtual supercomputer

Tue, 02 Dec 2008

For those ready to get their geek on, Purdue University has created the computer game for you.

Rack-A-Node is an online video game that lets those geeks who love both science and technology try their hand at designing and operating a simulated research supercomputer.

For the full story see: http://news.uns.purdue.edu/x/2008b/081110BowenGame.html

Purdue Cluster Challenge team all set for Texas

Tue, 02 Dec 2008

Face it, even Peyton Manning would have trouble hitting his receiver were the Colts quarterback standing in Indianapolis while his target ran in Massachusetts.

But long distance connection hasn’t been a problem for the Purdue SuperComputing ’08 Conference Cluster Challenge team.

Huddled over their laptops in a classroom lab on the Purdue Campus recently, Andy Howard, Alex Younts and colleagues were making adjustments to their competition machine—running in the Boston area at Purdue partner SiCortex—pretty much as if it were right in the room with them. They said they don’t even think much about the SiCortex SC1458 not being near enough to reach out and touch literally, outside of the occasional network lag that brings the point home.

Cluster Challenge teams at the premier international conference for high performance computing, networking, storage and analysis have to run a battery of benchmarking programs and working applications on their machines as efficiently as possible.

The students on the Purdue team—teams are limited to no more than six undergraduates—have spent most of the semester preparing to run a wide range of scientific software within the constraints of the contest, said Preston Smith, senior UNIX system administrator for Purdue's Rosen Center for Advanced Computing, who’s serving as team leader.

The Cluster Challenge’s rules also limit the amount of power the entries can draw from two 120-volt, 13-amp circuits, which limits the number of processors that can be employed. Purdue’s cluster had more than 40 processors last year, with most competitors falling between 30 and 60. The 2008 Boilermaker team may be able to run nearly 1,000 processing cores concurrently in the SiCortex, orders of magnitude above the competition.

“It’ll be markedly different than all the other teams,” said Younts, a sophomore in computer science from West Lafayette, who works as a student system administrator at the Rosen Center and was on last year’s Cluster Challenge team.

That’s because SiCortex builds supercomputers designed to deliver high performance using large numbers of slower, energy-efficient processors, both in power consumption and the cooling they require. Among other things, the company uses a unique, very fast “interconnect fabric”—the wiring that links its processors for working in concert—offsetting the raw speed disadvantage.

The Rosen Center, the research and discovery arm of Information Technology at Purdue (ITaP), the University’s central information technology organization, installed a top-of-the line SiCortex machine in June to test it from both performance and energy-saving perspectives. The SC1458 the Cluster Challenge team will use is the next step down from the Rosen Center’s SC5832. The numbers in both cases refer to the maximum number of processors the machines can contain.

Like the Rosen Center SiCortex, the competition machine comes in a modular package basically ready to plug in and run. That means this year’s Cluster Challenge team didn’t have to spend time building a cluster, which the 2007 team did. Most of the 2008 team’s work has involved retooling software to run on its entry in a way that takes maximum advantage of its voluminous processor array.

“It’s not your standard architecture so you can’t just install it and it works,” said team member Ryan Weinschenk, a senior in electrical and computer engineering technology from Noblesville. “The hardware performance is there. It’s just actually getting the software to run.” Mark Blessing, vice president of marketing for SiCortex, said getting a lot out of a thousand processors poses a challenge and the Purdue team has worked hard to tune the software to do it.

Team members anticipate being able to run many—and maybe all—of the applications concurrently, by parceling out a certain number of processors for each, which is potentially a big advantage.

“For the applications that we have running, I feel pretty confident,” Howard said.

The SiCortex also looks like it will easily beat the Cluster Challenge power limit, even when running the most demanding of the required benchmark software.

John Campbell, associate vice president for information technology, who heads the Rosen Center, said Purdue’s Cluster and Bandwidth challenge entries at SC08 are ultimately about pushing the limits of technology to meet the demands of the scientific community.

Even the theme of Purdue’s SC08 booth—No cycle left behind, no byte left unexplored—is a nod to the creative ways ITaP is finding to improve scientific productivity, Campbell said, including the testing of new technologies like the SiCortex that could offer significant benefits.

SiCortex has been impressed with the skill and enthusiasm of the Purdue Cluster Challenge team, Blessing said.

“By making the most of our computer’s unique architecture they are positioning themselves well for success in Austin,” he said. “More importantly, they will be able to apply what they have learned to the burning issue of supercomputing today: how to keep up with the ever-increasing demand for computing power in the face of severe data center power constraints.”

Smith said the Cluster Challenge team has been working almost since school started this fall and in some cases even before. Younts, Howard, a senior in electrical and computer engineering technology from West Lafayette, and David King, a senior in electrical and computer engineering technology from Lafayette, were able to work with some of the applications on the Rosen Center’s SiCortex over the summer. Howard is an assistant research programmer for the center and King is a student system administrator. Paul Willmann, a senior in computer technology from Carmel, is the other team member.

The students are enrolled in the high performance computing class of Jeffrey Evans, an assistant professor in the Department of Electrical and Computer Engineering Technology, as part of their involvement in the Cluster Challenge. But they spend up to 20 hours a week on the effort overall and have to juggle classes, homework and exams to attend SC08 for the competition. King will even be doing a little teaching.

He’s going to make a Skype call to his classmates in Evans’ computer architecture class and report on the happenings in Austin, Texas, site of the conference, which runs from Nov. 15-21.

After last year’s Cluster Challenge, Howard and Younts won the top student prize for a paper they presented at the Linux Cluster Institute conference April 29 to May 1 at the National Center for Supercomputing Applications in Urbana, Ill., and there are plans for a paper on this year’s experience, too, Evans said.

Meanwhile, SiCortex has now shipped the team’s machine to Texas for the competition, so the students will get to see it in person at last.

Writer: Greg Kline, (765) 494-8167, gkline@purdue.edu
Sources: John Campbell, associate vice president for information technology, (765) 494-1289, john-campbell@purdue.edu
Andy Howard, electrical and computer engineering technology, (765) 496-2434, ahoward@purdue.edu
Alex Younts, computer science, (765) 496-2023, ay@purdue.edu
David King, electrical and computer engineering technology, (765) 491-8158, dk@purdue.edu
Ryan Weinschenk, electrical and computer engineering technology, (317) 690-1906, cweinsch@purdue.edu
Paul Willmann, computer technology, (317) 496-2067, willmann.paul@gmail.com

Purdue Supercomputing '08 presence mixes projects to enable scientific productivity, education and fun

Tue, 02 Dec 2008

From an easy way to do cutting-edge science in a Web-based environment today to getting ready to take advantage of tomorrow’s next-generation high performance computers, presentations planned by Purdue researchers at the SuperComputing ’08 conference will cover a variety of topics in the field.

The presentations are part of the activity in Purdue’s booth at SC08, where visitors also will get a chance to build and manage their own supercomputer, virtually anyway. The SC Conference is the premier international gathering for high performance computing, networking, storage and analysis.

The booth at the world’s largest high performance computing conference is designed to promote Purdue, Information Technology at Purdue (ITaP), the University’s central information technology organization, and the Rosen Center for Advanced Computing, ITaP’s research and discovery arm. The conference takes place Nov. 15-21 in Austin, Texas.

The theme of the Purdue booth this year is: “No cycle left behind, no byte left unexplored.” That relates to the creative ways ITaP is finding to improve scientific productivity, said John Campbell, associate vice president for information technology, who heads the Rosen Center. For example, Purdue’s more than 20,000-processor strong Condor pool, a high-throughput distributed computing system that makes use of otherwise idle machines in offices, labs and elsewhere for research purposes, a project set to be highlighted in a colorful animated short film at the conference.

“We will be demonstrating a series of projects that we have implemented or expanded over the past year,” Campbell said.

Among other things, Purdue’s booth also will provide information to potential Purdue students and to job seekers about University academic programs and positions with ITaP and the Rosen Center.

The booth could attract some computer game fans as well. Purdue-designed Rack-A-Node is a “tower-defense” strategy game that has players build and operate a simulated supercomputer to manage waves of science jobs in fields ranging from climate-modeling and physics to chemistry and pharmacy, said Kyle Bowen, manager of ITaP’s informatics team, which created the game.

SC08 visitors will be able to give a console version of Rack-A-Node, which also is available on the Web, a spin at Purdue’s booth. Players start with a small computer and have to manage a series of jobs successfully to earn funding to buy an even larger machine and advance to subsequent levels of the game.

Purdue’s booth will feature two presentations on the HUBzero technology developed at the University, which makes it easy for researchers to create “electronic virtual organizations” for connecting with colleagues throughout the world and sharing ideas, computational resources and data. The technology also makes it possible to deliver in integrated fashion simulation tools, tutorials and workshops, podcasts and other resources via the Web, all in an manner so elegant, you have to see it to believe it, said Michael McLennan, senior research scientist and hub technology architect at Purdue and the Rosen Center.

Purdue has become a recognized leader in such “cyberinfrastructure” with the development of HUBzero, which powers nanoHUB.org and many other Web-based “hubs” for scientific collaboration.

At SC08, McLennan will give presentations on the HUBzero platform while Purdue electrical and computer engineering Professor Gerhard Klimeck will talk about nanoHUB, an international resource for nanotechnology theory, simulation and education that now has tens of thousands of users. NanoHUB is built on HUBzero.

Purdue presentations also will focus on taking advantage of the myriad processors available in high performance computing systems today and set to burgeon with the advent of petascale computers capable of more than a thousand trillion calculations per second.

• Purdue Professor Alex Pothen, head of Purdue’s Computing Research Institute, will outline recent results from the Combinatorial Scientific Computing & Petascale Simulations (CSCAPES) Institute.

• Faisal Saied, a senior research scientist at the Rosen Center for Advanced Computing and the Computing Research Institute, will discuss work toward petascale simulation at the Purdue Center for Prediction of Reliability, Integrity and Survivability of Microsystems (PRISM).

• Rudolf Eigenmann, professor of electrical and computer engineering and technical director of the Computing Research Institute, will cover methods for compiling programs to take advantage of today’s nearly ubiquitous multiprocessor systems (even laptops) and other high performance computing platforms.

• Other planned presentations will cover topics such as moving high-resolution scientific visualizations quickly over long-distance networks; using the Condor system to tap and combine unused compute cycles on server, office and lab machines for high-throughput high performance computing; accelerating applications with field programmable gate arrays, essentially chips that can be programmed to do specific tasks really well; and more.

More information:
www.rcac.purdue.edu/news/detail.cfm?NewsID=324
www.rcac.purdue.edu/sc08
www.rcac.purdue.edu/rackanode

Writer: Greg Kline, (765) 494-8167, gkline@purdue.edu

Purdue Bandwidth Challenge team: Go ahead, try this at home

Tue, 02 Dec 2008

Purdue’s Bandwidth Challenge entry at the SuperComputing ’08 conference is more like a NASCAR stock car than an Indy 500 racer.

Instead of a specialized design you would never see on the street, nor in your average data center, which Bandwidth Challenge entries and Indy cars tend to be, a team from the Rosen Center for Advanced Computing at Purdue built its system on a standard framework, albeit with some judicious tweaking. But while Purdue’s entry is based on open-source networking software and commodity hardware, the Purdue team thinks it still has plenty under the hood.

The Bandwidth Challenge is a standard feature of the premier international conference for high performance computing, networking, storage and analysis, taking place from Nov 15-21 in Austin, Texas. The competition pushes the limits on moving data over a long-distance computer network like the Internet or the TeraGrid, the National Science Foundation-funded, world’s largest network for open science computing, in which Purdue is a partner and resource provider.

The idea is that insights and innovations flowing from the nation’s scientists and their supercomputers are fed by torrents of data moving across continents on networks like the TeraGrid, the faster and the more consistently the better.

Moreover, the lessons learned could prove valuable in Purdue’s high performance computing operations, and for anyone else who wants to make use of them.

“What we have is open source, available to anybody right now if they wanted it,” said Ramon Williamson, the Rosen Center’s senior storage engineer, who’s serving as team manager. “We’re saying use your free stuff and tweak it and you can do this, too.”

The Rosen Center is the research and discovery arm of Information Technology at Purdue (ITaP), the University’s central information technology organization.

“Our Challenge entries are all about pushing the technology to meet the demands of the scientific community,” said John Campbell, associate vice president for information technology, who heads the Rosen Center. Even the theme of Purdue’s SC08 booth—No cycle left behind, no byte left unexplored—is a nod to the creative ways ITaP is finding to improve scientific productivity, Campbell said, like the project being highlighted in the Bandwidth Challenge.

Participants in the challenge are judged on the peak amount of data they can move from one place to another and how high a sustained rate they can maintain. Indiana University and partners won last year with a peak of 18.21 gigabits per second, in a pipeline with a maximum of 20, and a 16.2 Gbps sustained rate. At 1 Gbps, a decent-sized novel transfers in about a hundredth of a second.

This year, the competition also is placing a premium on “real-world applications and data movement issues” in addition to merely filling the pipeline, however.

That got Williamson thinking about the BlueArc servers the Rosen Center recently began using to manage data traffic. The cost-effective “appliances” run field programmable gate arrays (or FPGAs), in essence computer chips that can be reprogrammed for a specific purpose, rather than being hard-wired generalists. In BlueArc’s case that job is storage traffic cop. The tailored machines offer a considerable speed advantage, Williamson said.

The trick was to translate that into an advantage over a long-distance network using standard protocols like NFS (for network file system) and TCP (transmission control protocol), which has been the purview of Williamson’s colleague Michael Shuey, high performance computing technical architect at the Rosen Center.

“We see a lot of researchers needing to move a large amount of data across the country,” Shuey said. “We thought maybe we could tune NFS to do this.”

One reason they thought that: they couldn’t find anybody who had tried since the 1990s, when networks were considerably slower and network technology less advanced.

Shuey said writing transmitted data to storage in a manner that keeps the flow clipping along wasn’t much of a problem, but reading data from storage efficiently has been a challenge. The protocols read ahead to try to keep the pipe full, but they are made for moving material from local storage actually in the computer, such as a hard disk, or from nearby within a data center. They don’t account for the time lag in a cross-country transmission, which Shuey has had to find a way around, basically by getting the system to stock up on more data in the reading-ahead process than it would normally.

Greg Veldman, storage administrator for the Rosen Center, and Patrick Finnegan, UNIX system administrator, are the other members of the team. The team is partnering with BlueArc, Texas Memory Systems, makers of fast solid-state storage units akin to a USB flash drive, and Foundry Networks, which is providing the high-speed network connections between the storage appliances and the SC08 network in Austin for the competition.

James Reaney, BlueArc’s director for research markets, said the company likes working with partners who can apply its technology to real-world problems.

“I'm looking for research customers who have both the technical expertise and the willingness to push the boundaries of storage solutions,” he said, “and that's exactly why we partnered with Purdue for their entry into the Bandwidth Challenge event.”

If successful, the demonstration may show that researchers can work effectively with their large data collections not only close to home, “but perhaps they can access, manipulate and manage all that research data from just about anywhere,” Reaney said.

The plan is to move data—for climate simulations or something analogous—from the competition storage system in Purdue’s SC08 booth in Austin to West Lafayette for processing on the Rosen Center’s Steele cluster. Meanwhile, what the Rosen Center is learning from the exercise, besides being of value to other high performance computing centers, should prove useful at home, Williamson said.

A cluster to be built at Purdue Calumet from parts of West Lafayette’s decommissioned Lear supercomputer, which was displaced by Steele, is likely to be accessing storage on the West Lafayette campus and could employ something like the system developed by the Bandwidth Challenge team.

In addition, at some point the Rosen Center, pressed for space in its machine room in the Mathematical Sciences Building, might create a central, remote storage facility in the West Lafayette area or elsewhere, which could function, using the techniques, as if it were on site.

Writer: Greg Kline, (765) 494-8167, gkline@purdue.edu
Sources: Ramon Williamson, (765) 496-1272, ramonw@purdue.edu
Mike Shuey, (765) 494-0985, shuey@purdue.edu

Purdue project aims to put isotope analysis on the map, and the Web

Thu, 11 Sep 2008

Researchers at Purdue, its Earth and Atmospheric Sciences, Library and Statistics departments and the University’s Rosen Center for Advanced Computing are aiming to make it easer for ecologists, environmental scientists, hydrologists and many others, including policy-makers, to use the isotope data for a variety of purposes.

They’re working on a cyberinfrastructure for tapping an extensive, distributed data collection on line, and related Web-based tools allowing real science to be done with it, under a three-year, $831,573 grant from the National Science Foundation, its Division of Biological Infrastructure and the National Ecological Observatory Network.

The system focused on modeling stable isotope data with respect to space and time also will take advantage of the computational and storage resources available through the TeraGrid, the world's largest open science computing network, which is funded by the NSF as well. Purdue through the Rosen Center is a TeraGrid partner and resource provider.

“You need to make it easy for people to use (the isotope data) and incorporate it in their own research,” said Purdue earth and atmospheric sciences Professor Gabriel Bowen, who employs such data in his own research on large-scale, regional to global environmental change. He is a member of the Purdue Climate Change Research Center and co-directs the Purdue Stable Isotope Facility.

The project will include online tutorials to facilitate data access and use by scientists and non-scientists alike.

Water being the stuff of life, and hydrogen and oxygen the stuff of water, that makes hydrogen and oxygen isotopes a handy, useful and near ubiquitous avenue for the research Bowen conducts, which includes potential forensic uses as well as environmental science, and a wide range of other science and policy applications.

The mix of isotopes—atoms with variances in their number of neutrons—varies according to where those atoms have been, allowing someone like Bowen, for example, to trace the water coming from a tap in a town to snow melt from a mountain peak by reading, in effect, its isotope fingerprint. Likewise, water isotope tracers can be used in similar fashion to determine the geospatial origins of biological and geological materials.

In a sense, INPort, it stands for Isotopes Network Portal, will link a large range of scientists and non-scientists transparently to a vast fingerprint file of isotope data with integrated querying and geospatial modeling operations akin to a Geographic Information System, or GIS. Among other things, it will be able to produce maps showing the distribution of isotopes geographically and over time.

Besides Bowen and Lan Zhao, a Rosen Center research scientist, principal investigators on the project include statistics Professor Tonglin Zhang, library science Professor Christopher Miller, a GIS specialist, and Jason West, an ecosystem science and management professor at Texas A&M University, with whom Bowen worked when the two were postdocs.

Bowen and colleagues, in addition to drawing on 40 years of rainwater and snowfall measurements collected at several hundred sites worldwide by the International Atomic Energy Agency and the World Meteorological Organization, already have put together a large database of regional isotope observations and spatial statistical analysis tools people are using for teaching, research and regulatory purposes.

When he began looking at expanding on the effort and making the data more accessible and easier to use, someone recommended Zhao and the Rosen Center’s scientific solutions group led by Carol Song, which has done similar science “gateway” projects, for instance a Web-based climate modeling system for TeraGrid users. Their approved proposal started in August 2008. Bowen said the plan is for an initial deployment of INPort within the next year.

Zhao said the intent also is to leverage her group's work in serving earth science data to researchers by creating software components that can be used with existing systems and for future projects with similar requirements in biosciences, geosciences and other fields.

The Rosen Center is the research and discovery computing arm of Information Technology at Purdue (ITaP), the University’s central information technology organization.

Writer: Greg Kline, 765-494-8167, gkline@purdue.edu

Sources: Gabriel Bowen, 765-496-9344, gabe@purdue.edu

Lan Zhao: 765-496-2079, lanzhao@purdue.edu

More information: waterisotopes.org

www.rcac.purdue.edu

"Green" scales well, according to Purdue's early users

Tue, 09 Sep 2008

Early “tire kicking” by Purdue researchers is winning the University’s new “green” supercomputer some fans, although the results have been mixed for others and they all caution that more refining and testing need to be done.

“It allows us to efficiently perform simulations much larger than we could do with other machines,” materials engineering Professor Alejandro Strachan said of the computer from Massachusetts-based SiCortex. “I think it’s a great machine to do consistently large systems.”

Purdue’s latest supercomputer won’t make any lists of the fastest supercomputers in the world, the country, Midwest, or state. It’s not even the fastest on campus.

But the machine from SiCortex has advantages that could make it a boon for some researchers because it can tackle big problems by breaking them into pieces and processing those simultaneously with a high rate of efficiency.

Moreover, the new supercomputer—which has more than 3,000 processors, kind of the brain of a computer—isn’t green only if it is running climate simulations, one potential use, at Purdue’s Rosen Center for Advanced Computing. SiCortex designed it to consume less power than conventional supercomputers.

Purdue is the first university to test SiCortex’s top-of-the-line model. The high performance, research-focused computing arm of Information Technology at Purdue (ITaP), the Rosen Center is leasing the experimental machine for more than a year’s testing with an option to buy if it proves useful to Purdue researchers.

The project is co-funded by the Purdue Provost’s Office under a grant request supported by more than 50 faculty members, according to Gerry McCartney, Purdue’s vice president for information technology and chief information officer, who heads ITaP.

About a million-dollar machine, the SiCortex is low power in large part because the speed of its processors at 500 megahertz is a quarter to a sixth that of conventional processors, which reduces both power demand and heat, potentially lowering energy consumption for cooling and cooling costs as well. Its processors draw 600 milliwatts of power each, about the same as a cell phone or small flashlight. A standard supercomputer contains thousands of processors that require about 25 watts (25,000 milliwatts) each.

SiCortex makes up for some of the chip performance deficit with very fast communications among its thousands of processors. The machine also is built to move data quickly from storage to processing. That means its processors don’t, in effect, stand around waiting for something to do, like a painting crew out of paint.

So far, about 20 researchers and a dozen Rosen Center staff members have tried the machine and the verdict, in general, has been that some interesting things can be done with it, said William Whitson, ITaP and Rosen Center research computing manager.

Since its arrival in June, the SiCortex SC5832 at the Rosen Center has shown good scalability, which is to be expected given the communications-to-computation rate its speedy “interconnect fabric”—the communications pathways among its processors—and many processors permit, said Rudolf Eigenmann, professor of electrical and computer engineering and technical director of Purdue’s Computing Research Institute. That means if it takes an hour to simulate a sample of a certain size, researchers can double the size of the sample and double the number of processors in use and still get done in an hour, or very near to it, with a slight loss to overhead.

What its testers haven’t seen yet is an improvement in “absolute execution time,” which is to say the case where the SiCortex is significantly faster than a more conventional machine, Eigenmann said. But that could come with more exploration.

“Every new machine needs to be explored in how it can deliver performance the best,” said Eigenmann, one of the drivers, with McCartney, in bringing the new supercomputer to Purdue. The Computing Research Institute partnered with the Rosen Center on the funded proposal that purchased the machine and their staffs have worked together in helping researchers run and improve their computations on it.

In addition, Whitson said the Rosen Center plans to use a power-monitoring system on the SiCortex to examine it from the standpoint of energy efficiency eventually, since that’s one of the machine’s selling points.

Mechanical engineering Professor John Abraham focuses on improving efficiency in a different arena—internal combustion engines. His lab is on the lookout for things the industry could use to get gas and diesel engines to burn less fuel, improving mileage and also reducing emissions of climate-impacting greenhouse gases like carbon dioxide.

Abraham and colleagues also look for ways to reduce engine emissions of toxic pollutants, such as carbon monoxide, particulate matter and nitric oxide.

In addition, they study after-treatment systems, to catch and filter emissions over and above improvements in the combustion process, and fuel reformers, used to separate hydrogen from primary fuels for use in after-treatment devices and fuel cells.

All of this involves examining the fundamentals of flows, of fuel and oxygen moving turbulently through an engine system and reacting to generate the explosive power that drives a car or truck. Abraham’s lab looks at elements of the physical and chemical processes involved at micron scales in engines where the geometry scales are on the order of centimeters, which makes for a lot of microns to track. The researchers also resolve time scales on the order of nanoseconds or microseconds during an engine cycle that lasts for several seconds, once more a lot of individual elements to follow.

The complexity in the length and time scales, and the inherently multiscale nature of the problem, makes for models that require high performance computing. Abraham routinely uses resources at the National Center for Supercomputing Applications in Urbana, Ill., among others.

Meanwhile, Strachan specializes in molecular- and atomic-level simulation of materials in which he tracks what each and every atom in a sample of material is doing under various conditions.

While that sounds quite different from what Abraham does, the two Purdue research efforts share some important characteristics where their computing needs, and the Rosen Center’s SiCortex, are concerned.

The problems in both cases can be highly parallelized, that is they’re capable of being broken up into many smaller pieces and parceled out to a lot of computer processors. The processors then solve the pieces of the puzzle and reassemble the answers into a master answer. This can allow problems to be solved faster, larger problems to be tackled in a reasonable amount of time, or a combination of both.

With more than 3,000 processors, the SiCortex naturally interested Abraham’s and Strachan’s labs, along with the specialized interconnect fabric the company has developed. Interprocessor communications are important in what Abraham and Strachan do since they spread so many pieces among so many processors, which need to communicate to coordinate their activity.

“If you don’t have very fast communications between the processors it ends up killing you,” Strachan said. “That made it very attractive to us.”

Strachan and colleagues employ LAMMPS, widely used molecular dynamics simulation software developed at Sandia National Laboratories. In simulations looking at the mechanical response of a polymer material, and the underlying molecular dynamics that govern it to the breaking point, their results were encouraging.

Strachan said they found that their code scaled close to perfectly up to the machine’s full compliment of 3,240 processors. Other supercomputing resources available at Purdue can certainly handle the problems, Strachan said, and perhaps do it faster, but not as efficiently. They lose more time to overhead as the problem scales up.

Abraham’s students have run models of flow and chemical reactions on the SiCortex, using 128, 256 and 1,024 processors. Their code ported easily to the new machine and scaled well at the 128- and 256-procesor levels.

It didn’t scale as well on 1,024 processors, but Abraham said that’s probably more a matter of refining to use so many processors than a problem with the SiCortex.

“I’d like to use four times that many,” he said. “SiCortex is something attractive for us and promising so far.”

Mechanical engineering Professor Steve Frankel experienced similar scaling results—and better at the high end—in testing the SiCortex. Frankel is a computational fluid dynamics expert who focuses on turbulent flows involving everything from jet engines to human speech using computer modeling and simulation.

He primarily uses direct numerical simulation, or DNS, and large eddy simulation, or LES, methods to study turbulent flows, which Abraham also employs. The former provides basically a perfect rendering of a flow, while the latter confines the picture to its important elements but in finer detail and with simulation to fill in the gaps.

In either case, the number of points in a flow and number of time steps Frankel’s work incorporates require high performance computing to avoid waiting days, weeks, months, or years for results.

He and his lab labor to parallelize their code and make it scalable in running on their own and Purdue clusters, along with external resources like IBM’s Blue Gene supercomputers, on which Frankel in one test made use of 16,000 processing cores.

Frankel and colleagues also have used more than 2,000 cores for a human blood flow simulation on the IBM Blue Gene/L system at Argonne National Lab. His partnership with the national lab led him to try the SiCortex computer when it arrived at Purdue. Argonne acquired a SiCortex SC5832 in October 2007 and tests there showed similar performance under many conditions to the lab’s Blue Gene.

Frankel’s code runs best on a lot of cores—his lab used up 1,024 in the SiCortex—and with dedicated space to run continually. Under those conditions, they got scaling performance comparable to a Blue Gene, he said, just like Argonne.

Purdue aeronautical and astronautical engineering Professor Greg Blaisdell’s experience with the SiCortex is more mixed. Blaisdell also looks at turbulent flows, as they pertain to issues like the noise from jet aircraft and the drag and heat on aircraft wings, especially at supersonic speeds. Blaisdell’s lab doesn’t do it, but he noted that similar methods are employed to study gas flows on a galactic scale.

Blaisdell regularly makes use of Big Ben at the Pittsburgh Supercomputing Center, a more than 21-teraflop Cray XT3 system. He was intrigued by the 3,240-core SiCortex and he and his students wanted to test a reworked version of their code on as many of those processors as possible

“He was really pleased at how easy it was to get things up and running,” Blaisdell said of his student Phoi-Tack (Charlie) Lew, who had a job running the same day he got access to the system.

Blaisdell, Lew, aeronautical and astronautical engineering Professor Tasos Lyrintzis and graduate student Shih-Chieh Lo compared two large eddy simulations of a turbulent jet running on the SiCortex and Big Ben.

The SiCortex was slightly better as they scaled up from a minimum of 16 cores to 128, the limit for the test problem involved. Big Ben performed better in testing a second problem on up to 1,024 cores.

Blaisdell said testing on the SiCortex should help refine their code to better use the thousands of processors set to become standard with the advent of petascale computing, which Eigenmann said was one of the primary motivations behind obtaining the new machine.

The Purdue researchers agreed that being able to take advantage of more—rather than faster, as has been the case up to now—processors is likely to become the path to doing bigger science and engineering.

“That, I think, is the future,” Abraham said. “Individual processor speeds are not necessarily going up as fast as we want them to.”

Writer: Greg Kline, 765-494-8167

Sources: Greg Blaisdell, 765-494-1490, blaisdel@purdue.edu

Steve Frankel, 765-494-1507, frankel@purdue.edu

Alejandro Strachan, 765-496-3551, strachan@purdue.edu

John Abraham, 765-494-1505, jabraham@purdue.edu

William Whitson, 765-496-8227, whitson@purdue.edu

High Performance Computing School Set for September

Wed, 20 Aug 2008

A two-day high performance computing workshop for graduate students, postdocs, faculty and other interested researchers from on and off campus will be offered in September with support from five high-tech organizations at Purdue.

The Purdue School on High Performance Parallel Scientific Computing will take place Sept. 4 and 5 at the Burton D. Morgan Center for Entrepreneurship in Purdue’s Discovery Park, 1201 West State St., West Lafayette, IN, 47907-2057, and the Information Technology at Purdue (ITaP) computer lab in the Engineering Administration Building, 400 Centennial Mall Drive, West Lafayette, IN, 47907-2016.

The event is cosponsored by the Rosen Center for Advanced Computing, ITaP’s research and discovery computing arm, the Computing Research Institute, the Center for the Prediction of Reliability, Integrity and Survivability of Microdevices (PRISM), the Network for Computational Nanotechnology (NCN), and the Computational Science and Engineering Program.

For more information and to register, visit www.cri.purdue.edu/workshops.cfm.

Thursday morning’s session will present participants with a big-picture overview of what can be achieved with high performance computing. It features internationally recognized leaders in the field, said Faisal Saied, a senior research scientist at the Rosen Center, who’s leading the organization of the school with Purdue Professor Alejandro Strachan.

Strachan, assistant professor of materials engineering, said the session is designed to give participants “a glimpse of what is the state of the art today.”

Invited speakers include Steve Plimpton of Sandia National Laboratories, known for his work with the widely used LAMMPS molecular dynamics simulation software, and Andrew Lumsdaine of Indiana University, a parallel programming researcher who’s developed the open source message passing interface Open MPI, a system that allows many computer processors to communicate with each other and work in concert.

Other speakers will be Blaise Barney, a senior high performance computing researcher at Lawrence Livermore National Laboratory; Purdue mechanical engineering Professor Steve Frankel, an expert in computational fluid dynamics; and Purdue computer science Professor Ananth Grama, whose research focuses on parallel and distributed computing. Parallel computing parcels problems over more than one processor—thousands of which can be available in high performance computing—to solve bigger problems faster, from tricks viruses use to invade cells to secrets of the origins of the universe.

Thursday afternoon and Friday, school participants will get hands-on experience working with high performance computing hardware and software, Saied and Strachan said.

“That part will be (taught) primarily by Rosen Center staff,” Saied said. “Thursday afternoon and Friday are real how-to.”

Strachan said the school has a slightly different focus than previous Rosen Center high performance computing workshops. He said the September session is geared in particular to getting Purdue graduate students and post-doctoral researchers ready to work for the PRISM center and other research operations at Purdue as the school year begins, with an emphasis on making efficient use of high performance computing resources.

Strachan said most of the presentations will be recorded and made available on nanoHUB, www.nanohub.org.

The Rosen Center for Advanced Computing (RCAC) is the discovery and research arm of Information Technology at Purdue (ITaP). RCAC supplies high performance computing systems and storage—reliably and securely—for measured and computed data to faculty and staff doing computationally intensive cutting-edge science, engineering, and social science research.

The Center for Prediction of Reliability, Integrity and Survivability of Microsystems at Purdue, or PRISM, funded by the National Nuclear Security Administration, focuses on the behavior and reliability of microelectromechanical systems (MEMS), particularly miniature switches, developing advanced simulations for defense and commercial purposes.

The Network for Computational Nanotechnology uses advanced theory and simulations to explore new ideas for nanoscale devices, such as innovative types of transistors, that promise to help create future electronics. It operates nanoHUB, a Web-based toolbox and community for nanotechnology research.

The Computing Research Institute facilitates research in high performance computing at Purdue developing technology that realizes computational simulations to predict climate change, design new drugs, and explore the building blocks of matter, among many other things.

The Computational Science and Engineering Program is an interdisciplinary graduate program offering specializations in computational science and computational engineering for graduate students in the participating departments across the colleges of Science, Engineering, Pharmacy, Agriculture, Liberal Arts, and Technology. Specializations are offered at both master’s and doctoral levels.

Writer: Greg Kline. 765-494-8167, kline@purdue.edu.

Sources: Faisal Saied, 765-494-1583, fsaied@purdue.edu

Alejandro Strachan, 765-496-3551, strachan@purdue.edu

Strange molecule in the sky cleans acid rain, scientists discover

Wed, 13 Aug 2008

Researchers have discovered an unusual molecule that is essential to the atmosphere's ability to break down pollutants, especially the compounds that cause acid rain.

It's the unusual chemistry facilitated by this molecule, however, that will attract the most attention from scientists.

Quantum computing breakthrough arises from unknown molecule

Wed, 13 Aug 2008

The odd behavior of a molecule in an experimental silicon computer chip has led to a discovery that opens the door to quantum computing in semiconductors.

In a Nature Physics journal paper currently online, the researchers describe how they have created a new, hybrid molecule in which its quantum state can be intentionally manipulated - a required step in the building of quantum computers.

nanoHUB usage extensive

Wed, 13 Aug 2008

The Network for Computational Nanotechnology (NCN) mission is to support the National Nanotechnology Initiative (NNI) by designing, constructing, deploying, and operating www.nanoHUB.org, a national cyber-resource for nanotechnology theory, simulation, and education. The vision is to accelerate the transformation of nanoscience to nanotechnology and to build the nation’s nanotechnology workforce. Launched in 2002 with $10.5 million from the National Science Foundation and more than $4 million in support from Purdue University, NCN has deployed to date more than 90 simulation tools and 1,000 educational items authored by more than 500 members of the nanotechnology community. In 2007, NSF awarded NCN $18 million for a further five years of development and operation. From a web browser, researchers and students learn from nanoHUB educational materials, run state-of-the-art simulation tools, and view and interact with graphical results. In the 12 months ending June 30, 2008, more than 72,000 used nanoHUB. More than 265 research publications cite nanoHUB tools and/or resources, including a paper describing a new design for computer memory circuits and one analyzing a silicon nanowire transistor. During the 2007-08 academic year, 44 classes at 18 universities used nanoHUB resources. All Top 50 U.S. engineering schools and 14 percent of all .edu institutions in the U.S. use nanoHUB.

Shown below are the www.nanoHUB.org user locations in the U.S. on a Google map. nanoHUB has users at 14 percent of U.S. .edu institutions and all Top 50 U.S. engineering schools (U.S. News & World Report rankings).

Class papers become prize winners for supercomputer-building Purdue students

Wed, 13 Aug 2008

For two students in Jeffrey Evans’ high-performance computing class at Purdue, homework turned out to have an unintentional, and lucrative, benefit.

Andy Howard and Alex Younts, both of West Lafayette, Ind., walked away with the top student prize for a paper they presented at the Linux Cluster Institute (LCI) conference April 29 to May 1 at the National Center for Supercomputing Applications in Urbana, Ill. They won the best student paper award, and a $600 prize.

“We weren’t expecting that,” said Howard of the award from the premier international organization on building and managing supercomputing clusters running the open-source Linux operating system. “We were pretty excited.”

Clustering, the way supercomputers predominantly are built now, involves linking basically off-the-shelf computers with powerful processors -- the kind found on a lot of desktops and in a lot of laps these days -- to work in concert on big tasks from studying the behavior of thousands of molecules in the body to studying millions of stars in the cosmos.

Howard and Younts, currently working for the Rosen Center for Advanced Computing at Purdue, were part of the Purdue SuperComputing 2007 Cluster Challenge team, which competed at SC07 in Reno, Nev., last fall. The conference is the international supercomputing community’s major annual gathering.

The students built their cluster in Purdue Professor Evans’ course from machines provided by partner HP. AMD and Matrix Integration also helped sponsor the project. The class is normally graduate level but was opened to undergraduates on the team, said Evans, an assistant professor in the Department of Electrical and Computer Engineering Technology.

Among the challenges presented by the competition rules: the student-built clusters had to run a variety of benchmarking and working applications as optimally as possible. The programs stressed features ranging from memory handling to the speed at which data moves through the pipeline. Team members became the resident expert for an application, in essence, and did a paper about it for class.

nbsp;

Purdue junior Howard, for example, concentrated on Parallel Ocean Program, a simulator covering factors in the behavior and health of the oceans, which weighs into studies of global warming and climate change. Younts worked with POV-Ray, a heavyweight 3-D graphics and animation creation program, the kind of thing used to create the special effects magic in summer blockbuster movies.

Evans encouraged Howard and Younts to combine their class papers into a presentation for the Linux Cluster Institute conference this spring. They used their experience with specific programs to discuss building and operating a cluster in general. Younts presented the paper.

“There were several graduate students there with different papers on various topics,” said Younts, a Purdue sophomore who actually started working for the Rosen Center in high school. “But I guess everyone liked our presentation best.&rdquo

Rendering environment accesses huge pool of resources

Wed, 13 Aug 2008

Rendering the frames in a complex animation can tie up a work station, or a whole lab of them in a computer graphics class setting, for literally days or weeks, says Purdue Professor Gary Bertoline. But if the load could be shared among hundreds, even thousands, of machines a job that takes, say, 1,000 seconds on one might be done in a second, give or take a few ticks of the clock for latency and overhead.

Enter the Distributed Rendering Environment, or DRE, developed at Purdue and now available to users of the TeraGrid through TeraDRE. Like Purdue's DRE before it, TeraDRE draws on the Purdue Condor pool, a system for managing and sharing unused compute time on more than 7,000 linked computers on campus. Eventually, the idea is to incorporate computing resources from partners around the country connected through the TeraGrid, the world's largest open science computing network, said Bertoline, distinguished professor of computer graphics and director of Purdue's Envision Center for Data Perceptualization.

"Now when you press the render button, you have a huge pool of resources out there," Bertoline said. "We've got this working quite well here. It's being used in classes on a regular basis. We're working on taking it to a national level."

TeraDRE works with popular 3D modeling, animation, effects and rendering software such as Maya and open-source Blender. Laura Arns, associate director and research scientist at the Envision Center, said the list should continue to expand. Arns said Purdue also is refining the system to make it more aware of connected computers with special graphics-handling capability, so it can focus jobs on those automatically where possible. Another goal is refining the TeraDRE interface for better ease of use by users often more artistically than computationally centered.

Purdue Professor Nicoletta Adamo-Villani and her students use the DRE system in computer graphics technology courses that, among other things, are creating a highly graphical educational computer game, called "Nano Factor," designed to teach junior high school students about micro and nano technologies. Most recently, the assistant professor of computer graphics technology used TeraDRE to create a virtual rendition of a proposed satellite city for housing Istanbul residents in the event of a catastrophic Earthquake, a project with colleague Mete Sozen, Purdue's Kettelhut Distinguished Professor of Structural Engineering.

Adamo-Villani, also an Envision Center research scientist, said students in her class who built an introductory movie for "Nano Factor," which brings to mind the Oscar-winning feature "Toy Story," ended up with jobs at Pixar, the film's maker, and other major animation houses. But the DRE system is employed for visualization in cutting-edge science as well, such as a look at how a virus lands on, attaches to and implants itself in a host, Arns noted for example. She said the load-sharing system allows scientists and other animators to refine what they're creating on the fly, too, without waiting hours, or days, or weeks.

"The advantage is always speed," Adamo-Villani said. "It's a great resource to have."

Students take top honors at TeraGrid conference

Wed, 13 Aug 2008

Sanjiv Kumar is something of a fortune teller, but he and his colleagues don’t use a crystal ball to peer into the future of water systems like the St. Joseph River Watershed in northern Indiana and southern Michigan.

Instead, they use supercomputers at Purdue University and also spread across the country on a high-speed network dedicated to cutting-edge science.

The researchers project how climate change and other factors may affect, for instance, flow and water quality. They’re also identifying management practices that may help keep a system like the St. Joseph, which supplies water for more than 1.5 million people in a 15-county region around South Bend, viable in coming years.

“This tool can be very well applied … to other watersheds,” said Kumar, whose research poster about it during the TeraGrid ‘08 conference June 9-13 in Las Vegas turned out to be an award winner.

Purdue representatives took three top honors in two competitions at the annual conference focusing on the world’s largest high performance computing network for open scientific research. Purdue, in an effort led by the Rosen Center for Advanced Computing’s Carol Song, is a TeraGrid partner, contributing computing resources and developing technology for use on the system. That includes largely Web-based portals, called “science gateways,” designed to make doing research easier.

Kumar, a doctoral student in civil engineering, said the water system modeling might take weeks to a year on even a powerful desktop computer, depending on the amount of data being fed it. The supercomputers accessible through the TeraGrid and on campus at Purdue can reduce that to hours, or less.

Besides Kumar, who won first place in the student research competition, Purdue student Vinaitheerthan Sundaram, a doctoral student in electrical and computer engineering, won second place in the same competition. Sundaram, who works for the Rosen Center, presented a system for visualizing data from the latest generation of Doppler radar, interactively and in 3-D. He developed the system with Purdue Professor Bedrich Benes and his graduate student Yi Ru in the Computer Graphics Technology Department of the College of Technology and with Rosen Center researchers Song and Lan Zhao.

That’s the same data from radar stations nationwide you use to decide whether to take along an umbrella in the morning. But the flat 2-D pictures and even the 3-D graphics on the Weather Channel don’t make use of all the information hiding in the radar data, Sundaram said.

Purdue researchers created tools to automatically collect, process and render the data in near real time resulting in an interactive 3-D picture that can allow observers not only to see a storm coming, but spin it around and look at it from various angles, peer inside and more. When the weather is really active, as with the recent rains that have flooded parts of Indiana and the rest of the Midwest, the volume of data the system has to deal with is huge. That’s where the high performance computing resources available on the TeraGrid come into play, especially Purdue’s Condor pool, Sundaram said.

Condor puts the unused compute cycles of networked machines–on desktops, in labs, and servers–to work when the computers otherwise would be idle. Purdue has one of the largest Condor pools for distributed computing in the world, employing more than 15,000 processors on campus. In 2007 alone, BoilerGrid, as the system is called, provided more than 10 million hours of computation time for researchers and students, at Purdue and elsewhere.

Speaking at the TeraGrid’s annual conference, the National Science Foundation’s Steve Meacham, senior science and technology advisor in the NSF Office of Cyberinfrastruture, said the agency will be looking at high throughput computing resources, like Condor, in funding the next round of TeraGrid expansion. Meacham’s office oversees the TeraGrid.

“The fact that NSF specifically lists high throughput computing in (its funding proposals request) shows the value our Condor pool brings to the TeraGrid and the need for such resources from the user community,” Song said.

As student competition finalists, Kumar and Sundaram received travel, housing and registration expenses related to the conference. A panel of judges evaluated the entries on scientific merit and potential for impact. There were divisions for high school, undergraduate and graduate students, with the Purdue students taking the top two spots in the latter.

In addition, Rosen Center Customer Service Manager Kay Hunt won first place in the general poster competition for her poster on the TeraGrid Campus Champions program. The new program provides training and support for local representatives who then serve as ambassadors for high performance computing and the TeraGrid on their campuses. The aim is to spread the use of the tools.

The conference was an opportunity for the initial group, located on campuses from Hawaii to New Jersey, to meet and served as a kickoff for the program, which began enlisting participants in March.

But Hunt, who organized Campus Champions from scratch in December, said the meetings drew a crowd beyond the program’s initial 11 participants. More than a dozen additional schools are now considering participation. “There was a lot of interest from other institutions that had heard about it and wanted to check it out,” Hunt said.

A full list of Purdue’s participation in the TeraGrid conference can be found at: http://www.rcac.purdue.edu/about/TG08_purdue_pubs.htm.

Song highlighted a field programmable gate array tutorial by Rosen Center staff members David Braun and David Matthews. They outlined how the technology for programming, or reprogramming, chips more or less on the fly, the better to accomplish specific tasks, can be used in scientific computation–accelerating it, sometimes radically, while also lowering power consumption to make for “greener” high performance computing, a recent emphasis at ITaP.

Braun was able to recruit new users to the Rosen Center’s Brutus resource, which Purdue makes available to develop field programmable gate array code, part of its TeraGrid commitment. NSF’s Meacham also said such experimental resources will be among the criteria for funding the network’s expansion.

Summer research interns get early look at graduate student life

Wed, 13 Aug 2008

When Purdue Terrestrial Observatory Director Gilbert Rochon gave a presentation recently about high tech mapping of Indiana’s flooding this summer, employing satellite imagery from the observatory, Jean Pierre Antelo and Kaiem Frink were there with him.

Their names, that is, as co-authors on the research poster Rochon, associate vice president for collaborative research for Information Technology at Purdue (ITaP), unveiled in Boston at an international symposium on geoscience and remote sensing.

Not bad for undergraduates who just completed their bachelor’s degrees and signed on to be research interns this summer at Purdue.

In addition to working on flood mapping, Antelo, who graduated from the University of Florida in May, worked with Purdue management information systems Associate Professor Jackie Rees. That project involved better understanding the need for, and impact of, information security policies in fledgling biotechnology firms.

“They have a lot at stake,” said Rees, whose research focuses in part on information security and security policy. “If you lose your intellectual property, really in any startup, you’re toast.”

Rees became a mentor in the summer internship program in response to a call from Rochon. She liked the program’s goals and thinks it’s important for faculty members to “develop the talent pool.” The experience has been beneficial for her as well.

“I’ve just had a wonderful time with it,” she said.

She and Antelo talked about his interests in arriving at the biotech information security project as a logical confluence with her research. He produced kind of a primer firms considering such policies could use. “He has done a nice job with pursing this,” said Rees, who originally tried to interest her graduate students in the project, without generating a lot of enthusiasm.

Antelo said the experience gives him an idea of what it would be like to pursue a doctorate, something he didn’t think was for him. Now, he’s leaning in the direction of graduate school, maybe even at Purdue. (His brother Jean Paul graduated from the Krannert School of Management in 2002.)

Jean Pierre Antelo could end up on another poster, if one like it is ever printed—hailing the intent of the internship program. The program is designed to encourage talented undergraduate students from social and economic backgrounds that often don’t push research careers to move onto graduate school, and to assist in their preparations for graduate study.

Five interns were sponsored by ITaP, the Purdue Terrestrial Observatory and the NASA-funded Indiana Space Grant Consortium based at Purdue under the Purdue Research Opportunities Program (PROP). The Purdue Graduate School hosts the students through its Summer Research Opportunities Program (SROP), which hosts interns sponsored by other campus organizations, too.

In all 37 undergraduate students from around the U.S. and Puerto Rico participated in SROP this summer. They worked in nearly two dozen science and social science research groups with more than 30 Purdue faculty members, according to Melissa Danner, the program’s coordinator in the graduate school.

Besides Antelo and Frink, who’s from Elizabeth City State University, Elizabeth City, N.C., the ITaP interns included Milton Flournoy IV, Martin University in Indianapolis; Julie Nelson, South Dakota State University; and Jerrard Smith-Hopkins, Dillard University, New Orleans. The students receive intensive research experience with Purdue faculty mentors as well as produce research poster and oral presentations. They also attend graduate school admissions test preparation workshops, evening sessions on career opportunities for PhDs in various fields, and more. They take Graduate Record Examination practice tests both before and after their preparation classes.

“It’s nice to know what to expect,” said Nelson, who will be a junior at South Dakota State this fall and is considering pharmacy school. Some of the lessons the students learn aren’t academic in nature. Nelson said she cooked for herself, and grocery shopped, for the first time as a result of her internship. (The interns stay in Purdue’s Hilltop Apartments student complex.)

Smith-Hopkins, whose New Orleans school has around 1,000 students, said he would like to see Purdue, with an enrollment 40 times larger than Dillard, when all of its students are here.

“Being on a campus this big, walking 30 minutes and you’re still on campus, that’s wild,” he said.

The internship comes with a stipend and travel and housing expenses. Rochon, chief scientist for ITaP’s Rosen Center for Advanced Computing in addition to his other duties, said Intel and Motorola have helped ITaP sponsor its interns previously. The Indiana Space Grant Consortium agreed to fill in this year.

Frink actually started working with remote sensing and Geographic Information Systems as a freshman at Elizabeth City, including one program tracking conditions in the Antarctic using satellite imagery. He likes the idea of a career in GIS-related research.

“There are so many opportunities,” he said. “You can help people from all over the world you may never even meet, through a computer, through software.”

One of his professors at Elizabeth City met Rochon at a conference in Kenya and pointed Frink to ITaP’s summer internships. Frink said he knew graduate school was in his future a long time ago. "This gives me a foundation,” he said. “Purdue’s definitely in my top three.”

Danner said 79 participants have gone on to earn their doctorates since the SROP program began in 1980, 41 of them at Purdue.

Purdue graduate student Cliff Robinson, a former PROP student who earned his bachelor’s degree at Texas Southern University in Houston, will finish his master’s in agricultural and biological engineering in August.

For graduate school, he wanted to try a different kind of campus than Texas Southern. Purdue, as a big Big Ten university in the rural Midwest, fit the description. It also struck him as having “a really strong academic atmosphere.”

Robinson, another of the co-authors on the flood-mapping poster Rochon presented in Boston, took GIS courses offered by the Agricultural and Biological Engineering Department. He ended up wanting to be a GIS analyst. He’s always liked working with computers, but in an applications rather than programming or computer science theory way, he said.

“Like the FEMA flood maps, you’re actually able to give them a product right away,” he said. Robinson, along with Frink and Antelo, participated in a Purdue project this summer to create computerized, GIS-based 100-year flood maps for 785 U.S. counties under a contract with the Federal Emergency Management Agency. Rochon was co-principal investigator on the project.

Rochon also created opportunities for Antelo, Frink, Robinson, and Smith-Hopkins, to be co-authors on a research presentation about overcoming bandwidth and satellite communications limitations in remote sensing and high performance computing applications for African development. That presentation also was made at the Boston conference and Rochon plans to use it in the fall 2008 conference of the African Association of Remote Sensing of Environment, to be held in Accra, Ghana. Flournoy ended up a co-author on a presentation for that conference as well, addressing early warning and mitigation of vector-borne and zoonotic diseases.

Flournoy, a Martin University biology graduate who’s from Indianapolis, said he would like to work in business or industry for at least a year before continuing his education. He’s considering medical school after. The experience in the Purdue program has him reconsidering graduate school and a research career, however.

The summer internship program matched him with chemical engineering Professor Michael Harris. Flournoy’s research for Harris centered on improved sorting of nanoparticles ranging from 200 to 10 nanometers in size. The particles might have a variety of uses, from nanoelectronics, on which Harris and his lab work, to medical applications as miniscule delivery vessels for, say, targeted cancer treatment.

“I find the research that I’m doing very important,” Flournoy said. “To me it’s cutting edge. It’s exciting because of the implications it has.” Harris, who’s also associate dean of engineering for undergraduate education, frequently has undergraduate students working in his lab. He said he tries to give them projects where they can be productive enough to contribute to future scientific publications or grant proposals. He likes them there for another reason as well.

“I’ve recruited some of my best graduate students in that way,” he said.

Purdue project for FEMA producing a flood of flood maps

Wed, 13 Aug 2008

Classify Kaiem Frink as a first-shift worker in a flood factory running on Purdue University’s campus this summer.

Frink and the rest of the student work force, two shifts of them a day and one on weekends, aren’t actually filling the labs they occupy in the Agricultural and Biological Engineering building with water, of course.

The floods they’re producing are on maps of 100-year flood plains in more than 700 U.S. counties, not old-fashioned maps on paper but in a computer as part of a sophisticated Geographic Information System for the Federal Emergency Management Agency (FEMA).

Frink, as he watches blue stream reaches render on screen over a brown and green map of a Minnesota county, talks about it in Zen-like terms. “It’s a team project,” said the computer science graduate of Elizabeth City State University in North Carolina, who’s at Purdue this summer, co-sponsored by Information Technology at Purdue (ITaP) and the NASA-funded Indiana Space Grant Consortium, in an internship program to encourage him to go on to graduate school. “You have some individuals you can converse with, bounce ideas off them. The day goes fast. It’s peaceful. At least it is to me.”

Nonetheless, the job is a big one and Purdue is doing the biggest part of it.

FEMA wants to create a set of standardized “apples-to-apples” 100-year flood analyses for all the non-coastal counties in the U.S.

The federal emergency agency enlisted the Polis Center at Indiana University-Purdue University Indianapolis in the effort because of its expertise in using FEMA’s HAZUS-MH tool. HAZUS-MH is a GIS-based software application used to conduct scientifically grounded loss estimations from natural hazards such as floods, hurricanes, and earthquakes.

“It significantly exceeded our capacity,” Polis Center Associate Director Dave Coats said of the FEMA flood-mapping project.

But the Polis Center already works with Purdue and other universities in Indiana to assist local governments in employing GIS technology. The center was able to enlist the universities for the FEMA project as well. The Purdue Terrestrial Observatory, part of Purdue’s Rosen Center for Advanced Computing and ITaP, and the Department of Agricultural and Biological Engineering are spearheading the effort at Purdue.

“We completed Indiana four years ago with a separate grant,” said John Buechler, the Polis Center’s GIS manager. FEMA has about 10 states done overall and is gunning to finish the rest by September.

Purdue started with 300 counties. FEMA recently extended its contract to take in 485 more. Besides Minnesota, the list includes counties in Alabama, Arizona, California, Georgia, Idaho, Montana, Nevada, North Dakota, Oregon, South Dakota, Texas, Utah, Washington, and Wyoming. “We did all the big ones early in the process,” said Larry Theller, GIS specialist for the Purdue Agricultural and Biological Engineering Department. “We told them we could have 25 computers and that made us the biggest player.”

“This is a very large job that needed to be done in a short period of time,” said Purdue Professor Bernie Engel, who chairs the department and is the co-principal investigator on the FEMA project.

In a bit of kismet, Engel’s department, set to replace its computer labs in the fall, had a collection of new machines sitting idle over the summer.

Purdue also had a collection of more than a dozen staff members and students from the Agricultural and Biological Engineering Department, the Purdue Terrestrial Observatory and elsewhere ready for training sessions put on by the Polis Center to teach them to run a basic HAZUS-MH analysis. The 17-step process includes monitoring the software as it downloads and processes state geographic and flooding data, cleaning up variances between states and other data problems, and annotating the resulting map, in addition to dealing with the occasional software glitch.

“Some counties can be done in four hours, we’ve had some take two weeks,” said Larry Biehl, systems manager for the Purdue Terrestrial Observatory, who’s put in long hours as one of the staff members overseeing the FEMA project.

An international collection of student workers–from climatology, agricultural and biological engineering, civil engineering, urban planning, and other fields–works two shifts and on weekends to meet the FEMA deadline, said ITaP’s Associate Vice President for Collaborative Research Gilbert Rochon, director of the Terrestrial Observatory and co-principal investigator on the FEMA project.

HAZUS-MH is more than just a map, even a digital map. It epitomizes the idea of a Geographic Information System. For example, it incorporates data down to the census block level, allowing projections of everything from the number of displaced people and their ethnicity to the likely damage to buildings and the economic loss as a result of, in this case, a major flood.

In fact, the Polis Center and its partners are using the system to help a number of Indiana, Illinois, and Wisconsin counties get ahead of the game by fashioning pre-disaster mitigation plans, which FEMA wants every county to do.

The center’s partnership with Purdue is developing both research and international angles as well.

Biehl and colleagues at the Purdue Terrestrial Observatory are providing the satellite imagery that is the stock-in-trade of the facility to study the calculating of flood boundaries from the images based on this summer’s flooding in Indiana and around the Midwest. The plan is to compare that to on-the-spot field reports and see if such images can be employed in an enhanced GIS, Buechler said.

“Certainly, this has been a very timely issue with the floods the last few weeks,” Engel said.

Purdue researchers already have overlaid the flood images onto 2007 U.S. Department of Agriculture National Agricultural Statistics Service data to generate an estimate, 15 percent, of the corn and soybean crop submerged in Indiana’s flooded counties. Rochon presented the research at the Institute of Electrical and Electronics Engineers International Geoscience and Remote Sensing Symposium in Boston July 6-11.

Rochon said the partners also are looking at the possibility of doing flood mapping for the Nile River in Africa, particularly in Egypt, the Sudan and Ethiopia.

The Nile project could produce volumes of data such that it would take Rosen Center high performance computing resources like the new Steele cluster, with 812 dual quad-core compute nodes and a peak performance of 60 teraflops, the new low-power SiCortex supercomputer, or Purdue’s 14,000-processor Condor distributed computing pool to analyze the mass, Rochon said.

Buechler said bringing high performance computing into play also might allow the incorporation of real-time stream flow data and what-if modeling of floods.

Biehl said high performance computing could permit the kind of dynamic flood modeling FEMA would really like to do, incorporating rain data as it happens, for instance, and taking into account highly localized variances in rainfall within a county to quickly and accurately predict impacts downstream. The 100-year flood plain project is somewhat static by contrast, he said, providing in effect a snapshot of flooding in a worst-case scenario with an extraordinary rain countywide.

Purdue supercomputer unboxed and built by lunchtime

Tue, 06 May 2008

Staff members at Purdue University had hoped to build the Big Ten’s largest campus supercomputer in just a day on Monday, May 5.

But it didn’t take that long — they were done by lunch.

Purdue to install Big Ten's biggest campus computer in just a day

Sun, 04 May 2008

The largest supercomputer on a Big Ten campus will be installed at Purdue in a single-day, electronic "barn-raising."

More than 200 employees will gather May 5 to help build the massive machine, which will be about the size of a semitrailer when installed. It will be the largest Big Ten supercomputer that is not part of a national center.

Top CO2 emitting counties in the United States identified

Sun, 04 May 2008

The top twenty carbon dioxide-emitting counties in the United States have been identified by a research team led by Purdue University.

The top three counties include the cities of Houston, Los Angeles and Chicago.

Kevin Gurney, an assistant professor of earth and atmospheric science at Purdue University and leader of the carbon dioxide inventory project, which is called Vulcan, says the biggest surprise is that each region of the United States is included in the ranking.

Breath of a Nation -- Animated CO2 Map

Sun, 04 May 2008

Scientists have come up with a new way to precisely track daily and local patterns of carbon dioxide emissions from the burning of fossil fuels by power plants, factories, and vehicle traffic. The resulting database and maps provide a view of the “industrial metabolism” of our combustion-powered lives, Kevin Gurney, the leader in the project and an atmospheric scientist at Purdue, told me today.

Virus views enhanced using nation's largest Condor flock

Sun, 04 May 2008

A team lead by a Purdue University researcher has achieved images of a virus in detail two times greater than had previously been achieved.

Wen Jiang, an assistant professor of biological sciences, led a team that used the emerging technique of single-particle electron crymicroscopy to capture a three-dimensional image of a virus at a resolution of 4.5 angstroms, Approximately 1 million angstroms would equal the diameter of a human hair.

LSU HPC workshop via the Access Grid

Mon, 10 Mar 2008

The High Performance Computing group at Louisiana State University will be conducting an HPC workshop Tuesday and Wednesday, March 11 and 12. This workshop will be broadcast over the Access Grid and ITaP's Rosen Center has arranged to carry it from 9:30am-6pm each day in the YONG 511 conference room.

Information about the workshop, including a detailed agenda, may be found at

http://www.hpc.lsu.edu/training/20080311.index.php

Feel free to visit YONG 511 to attend any workshop sessions you're interested in. Please note, however, that the times given on the LSU web page are for the Central time zone rather than Eastern.

CMS Tier 2 Workshop to be held at Purdue -- April 11, 2008

Fri, 29 Feb 2008

The Rosen Center for Advanced Computing (RCAC) and Norbert Neumeister, who leads the CMS effort at Purdue University, are hosting the United State Spring Workshop for all of the Tier 2 sites participating in the Compact Muon Solenoid (CMS) experiment being conducted a the Large Hadron Collider at CERN.

Workshop information may be found here.

Proposed new city of Istanbul premiers in animated video

Wed, 09 Jan 2008

Istanbul is at such high risk for a devastating earthquake that engineers at Purdue University and the Republic of Turkey have come up with a bold new proposal: build a second city.

A second satellite city would provide immediate refuge to inhabitants of the old city in the event of a catastrophic earthquake and soften such an event's effects on the nation's economy.

Read More...

New specialization will focus on supercomputing

Thu, 20 Dec 2007

Purdue University is developing courses that will focus on high performance computing, making it one of a few universities nationwide to offer such curriculum.

The specialization wil be offered in the Department of Computer and Information Technology. It will include a course starting in the spring semester in which each student will build a high performance computer that is about four times more powerful than a standard personal computer.

Research forcasts increased chances for stormy weather

Mon, 10 Dec 2007

Researchers who study severe weather and climate change joined forces to study the effects of global warming on the number of severe storms in the future and discovered a dramatic increase in potential storm conditions for some parts of the United States.

House of Horrors

Mon, 29 Oct 2007

Visit the Envision Center Virutal Haunted House

Wednesday, October 31, 2007

1:00-5:00 PM

Envision Center

For more information conatct Amanda Hart at hart6@purdue.edu or 765-496-7888

Condor Boot Camp at Purdue - October 25, 2007

Wed, 10 Oct 2007

The Rosen Center for Advanced Computing (RCAC) at Purdue is developing a campus-wide and state-wide consortium tentatively called BoilerGrid. BoilerGrid will offer to participants the use of idle cycles from a large number of computers for the purpose of conducting scientific research.

Condor from U. of Wisconsin is a specialized workload management system for compute-intensive jobs. Condor provides job-queuing, scheduling policies, priorities, resource monitoring, and resource management. Condor is able to operate on idle cycles on regular computer hardware. The student lab that sits idle in your department for 12 hours a day can now become a compute resource useful to researchers around the state with "high-throughput" applications.

By harnessing these idle cycles on HPC clusters and student labs, RCAC is able to deliver millions of hours of computing time to researchers from Purdue, the state of Indiana, and the nation.

An especially valuable Condor application is the TeraDRE for rendering of animations. http://www.purdue.teragrid.org/content/view/48/25/

Condor “Boot Camp”
Thursday, October 25, 2007
10:00a.m.-3:00p.m.
Burton Morgan Center
Room 121
Purdue Campus

Audience:
Sysadmins from departments around campus and universities from the state of Indiana

Content:
Condor 101: the basics of Condor: what it does, and how to use it.

Basic Condor administration: Things you'll need to know to maintain your installation, configure execution policy on your machines, how to get help ...

Registration:

No charge. Lunch will be provided. Register to attend at http://www.itap.purdue.edu/training/registration/?offeringid=2303

Parking:

Parking permits will be available for non-Purdue attendees. Please pick up your permit between 9:30-10:00 am at the circle drive in front of the Burton Morgan Center.

Questions:

Technical: Preston Smith psmith@purdue.edu

Administrative (parking and other): Debbie Sterrett dcsterre@purdue.edu

RCAC system outage planned for October 5-6

Thu, 06 Sep 2007

All RCAC systems will be unavailable from 5pm Friday, October 5, thru Saturday evening, October 6, due to a scheduled power interruption in MATH. PBS jobs with walltimes greater than the time remaining until this outage will be held in the queue for execution after systems have been returned to service on October 6.

Please refer questions about this service interruption to rcac-help@purdue.edu.

Update on 8/20 RCAC system outage

Tue, 21 Aug 2007

System damage caused by the Monday, August 20, power failure has been repaired and all RCAC systems have been returned to service.

Please refer questions about this system outage to rcac-help@purdue.edu.

Scientists Create their own Web 2.0 network with nanoHUB

Tue, 21 Aug 2007

Teenagers may not have heard about it, but there's a Web 2.0 site that's a hit with scientists and engineers.

nanoHUB.org, a so-called science gateway for nano-science and nanotechnology housed at Purdue University, is taking the tools of Web 2.0 and applying them, along with a few tricks of its own, to further nano-scholarly pursuits.

Important news for Radon/Recycled cluster users

Thu, 16 Aug 2007

The Radon/Recycled Linux cluster was upgraded significantly during RCAC's August 15-17 maintenance window. Changes resulting from this upgrade include the following:

The operating system on radon.rcac.purdue.edu and the cluster's compute nodes was upgraded to the most recent version of Debian Linux -- Debian GNU/Linux 4.0 (a.k.a. etch).
The cluster's radon.rcac.purdue.edu front-end system was upgraded to newer hardware and its IP address was changed to 128.211.157.42.
The cluster's batch job scheduler was upgraded to PBSPro 8.0, which is the same job scheduler used on other RCAC Linux clusters.
Most compute nodes in the cluster were replaced with systems having faster processors and more memory. All nodes in the cluster now have at least 1GB of memory and 2.8, 3.0, or 3.2 GHz.
Due to the increased cpu speed on the cluster's compute nodes, the maximum walltime limit for PBS jobs has been set to 336 hours (i.e. 14 days).

Since the operating system upgrade resulted in changes to numerous run-time libraries, RCAC recommends that all Radon/Recycled cluster users re-compile their applications before running them in the cluster's new configuration.

Whenever possible, PBS jobs that were in the system prior to the upgrade were preserved and the owners of jobs that could not be preserved have been contacted via e-mail. However, due to the previously-mentioned run-time library changes, jobs that were preserved have been "held" and will not be automatically scheduled for execution when the cluster is returned to production.

These held jobs can be identified by running the "qstat" command and looking for jobs with an "H" in the job-status column. These jobs will also have job ids containing the string "xenon". Job submitted after the cluster's upgrade will job ids containing the string "argon".

If you see that you have held jobs you know would run programs you or members of your research group have compiled on radon.rcac, please re-compile your programs, re-submit your jobs, and use the "qdel" command to delete the held jobs. For example, if your job 12345.xenon is held, you could delete it by running the command

qdel 12345.xenon.rcac.purdue.edu@argon

If, on the other hand, you have held jobs you know are going to use RCAC-supplied software (e.g. Matlab or Gaussian), and you would like them to be scheduled for execution, you may use the "qrls" command to release their holds. For example, if you have a job 12345.xenon that is held, you could release it for execution by running the command

qrls -h u 12345.xenon.rcac.purdue.edu@argon

You may specify multiple job ids on the "qdel" and/or "qrls" command lines if you have several jobs you would like to either delete or release. Also note that you must specify the full job id including "xenon.rcac.purdue.edu" when deleting or releasing these jobs.

Any jobs that are still held on Friday, August 31, will be deleted by RCAC systems staff.

If you have questions about the Radon/Recycled cluster's upgrade, need help releasing held jobs or deleting jobs, or would like assistance re-compiling your programs, please contact the RCAC user support staff by sending e-mail to rcac-help@purdue.edu.

DXUL/Fortress maintenance planned for Wednesday, August 15

Thu, 09 Aug 2007

The DXUL/Fortress archival storage system will be unavailable from 9am-5pm Wednesday, August 15, while system maintenance is performed.

Please refer questions about this service interruption to rcac-help@purdue.edu.

HPC system maintenance planned for Wednesday thru Friday, August 15-17

Thu, 09 Aug 2007

The RCAC HPC systems will be unavailable Wednesday through Friday, August 15-17, so that system maintenance can be performed.

Systems involved include all RCAC Linux clusters, Condor, the SGI Altix, the IBM SP and Regattas, and the Sun F6800 cluster. Only PBS jobs with time limits short enough to ensure their completion prior to Wednesday morning, August 15, are being scheduled for execution -- PBS jobs with longer time limits will be held in the queue for execution after the maintenance period ends and the systems have been returned to service.

Please refer questions or comments about this service interruption to rcac-help@purdue.edu

Envision Center leads winning team in Google 3D Campus Contest

Thu, 19 Jul 2007

118 buildings of West Lafayette Campus are created. 3D Models include all academic and administrative buildings as well as a couple of parks and the major athletic facilities. Check out the Project Showcase here ! As part of the winnings, Google will fly and host Purdue team members at Google's Mountain View Campus.

Purdue makes distributed rendering available at SIGGRAPH 2007 via TeraGrid

Wed, 18 Jul 2007

Academic animators and researchers doing scientific visualizations have a new resource that will allow them to render their 3-D animations in a fraction of the time it can take on a single computer.

Attendees of the 2007 ACM SIGGRAPH conference in San Diego on August 4-9 can bring animation files with them and submit them for rendering at a Purdue University-sponsored "render farm".

'Push-Button' Climate Modeling Now Available

Thu, 12 Jul 2007

A tool used by scientists to create climate models is about to become easier to use and available to a much wider audience.

A new Web-enhanced version of the most commonly used climate modeling system will allow many more scientists -- and even curious students -- to test theories about the planet's climate.

Matt Huber, an assistant professor of earth and atmospheric sciences at Purdue Univeristy , says the Community Climate System Model is already used by thousands of scientists, and the results from their models often make headlines around the world.

Purdue offers Condor resources to TeraGrid community

Mon, 21 May 2007

Purdue University’s Rosen Center for Advanced Computing has become the largest provider of high-throughput computing cycles on the National Science Foundation’s TeraGrid.

Carol X. Song, senior research scientist in the Rosen Center and principal investigator for TeraGrid at Purdue, says that more than 6,400 computers of all sizes—from desktop machines used by students to do homework and check e-mail, up to large, powerful research computers—are linked together using the open source application Condor.

"By using Condor and making resources available over the TeraGrid, we are leveraging our national and international science resources," Song says. "We will continue to expand our Condor pool to include additional machines as well as machines at other campuses through regional grids."

The latest addition of more than 2,200 machines to the Condor pool includes 1,000 computer lab machines on the West Lafayette campus, 300 computer lab machines on the Purdue Calumet campus, 300 workstations at the University of Notre Dame campus, and 664 processors in a cluster maintained for research computing at the Rosen Center in West Lafayette.

Miron Livny, professor of computer science at the University of Wisconsin, says that Purdue's Condor pool is the largest in the nation.

“Purdue is committed to a vision, and they are making that vision a reality. I am pleased to say that early on I worked closely with people at Purdue, and we shared this vision for research computing," Livny says. “I think it's wonderful that Purdue has taken the leadership on this on the TeraGrid. And I don't pass out these kinds of compliments often."

One researcher, Michael Deem, Rice University's John W. Cox Professor of Chemical Engineering, has used more than 2 million hours of computer cycles at Purdue to catalog the chemical structure of compounds called zeolites. This team’s zeolite database now contains more than 3 million structures, and it is still growing.

Deem aims to identify and categorize as many of these structures as possible so that chemical engineers can select the exact zeolite they need. This is just the kind of high-throughput job that works well on Purdue's distributed computing system.

“The throughput is much higher there than I can get locally because of the large size of the Condor pool at Purdue," Deem says. “Purdue is doing a great service to the scientific community by providing this resource."

The distributed computing resource is available over the TeraGrid, of which Purdue is one of nine resource provider sites. Charlie Catlett, the chief information officer at Argonne National Laboratory and chair of the TeraGrid Resource Provider Forum, says that it is important to provide a variety of computing resources to researchers.

“High-throughput, or capacity, computing is extremely important to the TeraGrid user community," Catlett says. “Purdue and the Condor team have provided an excellent model for harnessing campus cyberinfrastructure in a way that benefits local users and also serves the national community."

The computers in the Condor pool at Purdue are used roughly 45 percent of the time for their intended purpose, 45 percent for Condor, and they are idle the other 10 percent of the time.

“This shows that our site can provide significant computing power to the nation without requiring dedicated resources," Song says.

In April, more than 5 million hours were awarded to users of Purdue Condor resources at the TeraGrid allocations meeting.

Preston Smith, a sysems research engineer for Purdue’s Rosen Center, says that Purdue has refined its use of the software by using it as a secondary scheduling system on the computers, which allows the computers to be put to use whenever they are available instead of making them available only at certain times, such as at night. The primary schedule for computing jobs at the Rosen Center is the Portable Batch System, or PBS. Purdue uses PBS Pro.

“The thing we do that is unique is that we use Condor in tandem with PBS Pro," Smith says. PBS Pro was developed by Altair Engineering.

Condor and PBS Pro are connected so that they can “talk" to each other before a job is assigned to see what computers are available. This scheduling tool allows Condor to send a job to a computer whenever it's not being used instead of at set times, which allows many more unused computing cycles to be harvested, Smith says.

Livny says that he hopes Condor usage increases at other universities and that the now-wasted cycles can be put to good use.

“Other campuses should follow Purdue's leadership," Livny says. “I believe this is the right way for us to move forward, get organized and get resources together, and then go out on the national level and share resources with other institutions."

Purdue’s Rosen Center for Advanced Computing publishes a daily graph showing Condor usage.

Proposal Deadlines Nearing

Fri, 01 Jun 2007

Department of Energy proposals and TeraGrid high performance computing resource allocation requests are due in the very near future.

Information regarding the DOE proposal schedule may be found at: http://www.supercomputingonline.com/article.php?sid=13748 .

Information about submitting proposals for the September xRAC meetings is available at http://www.ci-partnership.org/Allocations/allocations.html.

'Not so fast, supercomputers,' say software programmers

Fri, 01 Jun 2007

The fastest of the fastest computers - supercomputers used at national research centers, research universities and major corporations - will soon gain even more performance by taking advantage of multicore computing.

Despite the promise of almost unimagined computing power, however, even computing experts wonder whether this time the hardware developers have raced too far ahead of many programmers' ability to create software.

Atkins to Speak at TeraGrid '07

Fri, 01 Jun 2007

Speakers, Science Gateways and Tutorials -- Much in store for attendees at TeraGrid ’07!

CHICAGO — The second annual TeraGrid conference, being held June 4-8 in Madison, Wisconsin, will provide the scientific community with opportunities for knowledge sharing, collaboration and technologies to enable scientific discovery using the resources of the TeraGrid.

Registration is underway at http://www.teragrid.org/events/teragrid07.

Daniel E. Atkins, director of the National Science Foundation Office of Cyberinfrastructure, will speak at TeraGrid ’07. His talk, on the morning of June 6, is titled “The Ubiquitous TeraGrid: Everywhere but Nowhere.” Atkins chaired the National Science Foundation (NSF) Advisory Panel on Cyberinfrastructure that in 2003 issued the landmark “Atkins Report” recommending an Advanced Cyberinfrastructure Program to revolutionize science and engineering research and education. The recent NSF Cyberinfrastructure Council’s “Cyberinfrastructure Vision for 21st Century Discovery” builds on this theme.

“Dan Atkins is a leader in the vision of ubiquitous cyberinfrastructure and I look forward to this opportunity for the TeraGrid community to talk about how it contributes to that vision,” said Dane Skow, principal investigator of the TeraGrid’s Grid Infrastructure Group.

Previously announced speakers include keynote Anita K. Jones, who directed research at the Department of Defense from 1993-97 and managed a budget of $11 billion; Philip Maechling, who leads earthquake modeling at the Southern California Earthquake Center; and Paul Strong of eBay Research Labs, who directs strategy for infrastructure and enterprise management.

New to the speaker line-up is Ray Rose, co-founder and president of Rose & Smith Associates, a consulting group dedicated to sharing the principles of successful online learning and the innovative use of technology in education settings. “We’re very pleased to have such a prestigious set of speakers to share their insight into the role of cyberinfrastructure in accelerating science and commerce,” Skow continued. “We’ve seen a marked increase in new modes and scale of use of the TeraGrid this past year, as have our colleagues in commercial and social networking. The nearly 600 published papers this past year in almost 200 scientific journals from work done on TeraGrid are the first fruits of this investment.”

Conference attendees will also learn how to become involved in TeraGrid Education, Outreach, and Training (EOT) programs and how to build a TeraGrid Science Gateway. “TeraGrid’s year-long EOT programs provide training, workshops, institutes, and self-paced tutorials designed to increase knowledge and skills in cyberinfrastructure on scientific computing resources across all disciplines, and across all stages of learning from K-12 education through professional practice,” said Scott Lathrop, TeraGrid area director for EOT and External Relations.

TeraGrid Science Gateways enable communities of users with a common scientific goal to use national resources through a common interface. Currently, more than 20 science gateways from a wide variety of disciplines are using TeraGrid resources through science portals, desktop applications, and grids.

TeraGrid Area Director Nancy Wilkins-Diehr says that she is particularly excited about the full day tutorial — Building Blocks for a Simple TeraGrid Science Gateway — that will allow attendees to build a science gateway that accesses TeraGrid resources. “We hope this simple building block approach will help attendees envision how a gateway might extend the capabilities of their communities. Expert gateway developers will be on hand to talk about their work and answer questions.”

TeraGrid ‘07 will highlight the results of science enabled by TeraGrid resources. In addition, on Monday, June 4, thirteen tutorials on TeraGrid resources, such as visualization tools, science gateways, and Globus middleware will be offered. Many of the tutorials are intended for new and intermediate TeraGrid users who would like to broaden or deepen their use of TeraGrid resources. Currently, TeraGrid resources include more than 250 teraflops of computing capability and more than two petabytes of online and archival data storage, with rapid access and retrieval over high-performance networks. Researchers can also access more than 100 discipline-specific databases.

For more on TeraGrid and the TeraGrid ’07 conference, see http://www.teragrid.org.

About TeraGrid:

The TeraGrid, sponsored by the National Science Foundation Office of Cyberinfrastructure, is a partnership of people, resources, and services that enables discovery in U.S. science and engineering. Through coordinated policy, grid software, and high-performance network connections, the TeraGrid integrates a distributed set of high-capability computational, data-management and visualization resources to make research more productive. With Science Gateway collaborations and education programs, the TeraGrid also connects and broadens scientific communities.

TeraGrid '07 Conference Registration Now Open!!

Mon, 21 May 2007

Conference Registration is Open!

Registration is open for the second annual TeraGrid conference -- "TeraGrid '07: Broadening Participation in TeraGrid.�

The registration fee will increase after May 5, so register early! Please visit http://www.union.wisc.edu/teragrid07/registration.html for registration details.

Thank you to everyone who submitted papers for TG07! The paper submission process is now closed. You will be notified of paper acceptance by March 16.

Call for Posters, Demonstrations, BoF�s and Student Contests!

Don't delay! Submissions are being accepted for posters, demonstrations, BoF�s and Student Contests. For more information, please visit:http://www.union.wisc.edu/teragrid07/papers.html We encourage high school, undergraduate and graduate students to enter the student contests. To learn more about the three exciting contests, please visit: http://www.union.wisc.edu/teragrid07/contests.html

For general conference information, please visit: http://www.union.wisc.edu/teragrid07/

Meeting the Data Transfer Challenge

Fri, 02 Mar 2007

News about Purdue Tier-2 on the Data Transfer Challenge

Meeting the Data Transfer Challenge

Purdue teams up with California university to raise the bar in grid computing

Fri, 02 Mar 2007

Purdue University researchers have reached new milestones in grid interoperability through the successful integration of two Open Science Grid (OSG) sites running a scientific application over the National Science Foundation TeraGrid network.

Purdue-Industry HPC Workshop Registration Open

Fri, 02 Mar 2007

Purdue-Industry HPC Workshop

March 20, 2007

This workshop will bring together High Performance Computing (HPC) researchers at Purdue University and industry experts using parallel and high-performance computing technology. The goals of the workshop are to highlight existing uses of HPC in industry, to explore new possibilities for broadening this area to enhance economic competitiveness, and to identify new projects where Purdue can partner with industry in expanding the role of HPC.

Further information is available at http://www.cri.purdue.edu/industryHPCWorkshop.cfm

Envision Center wins honors from Campus Technology magazine

Mon, 11 Sep 2006

Purdue's Envision Center for Data Perceptualization was chosen by Campus Technology magazine as one of the winers of its Campus Technology Innovators 2006 competition.

Purdue won in the category of "Virtual and Immersive Learning". The awards were announced in the magazine's August 1st issue and is available at http://www.campus-technology.com/article.asp?id=18945

The article outlines Purdue's need for a way to display data in three dimensions; the faculty idea for the center; the steps of the grant, benchmarking, obtaining advice and donations from companies, and plans for the future. The article describes four of the technologies available in the Center: VR Theater, Motion Capture, Tiled Wall, and the Access Grid.

Campus Technology received 480 nominations. Purdue shared its category's top spot with Applachian State University which devised a scalable "world" for coursework, meetings, and social interaction.

Rosen Center Training Courses Available Now!

Fri, 18 Aug 2006

The Rosen Center training offerings for the Fall 2006 semester are open for registration.

Offerings include several introductory classes on RCAC computing resources, Linux clusters, the IBM SP, and the Sun F6800. Programming language classes include parallel programming with MPI, C++, FORTRAN, shell programming, and OpenMP as well as the use of the TotalView debugger. Other offerings include an introduction to the TeraGrid, an introduction to using the national computing centers, an introduction to using UNIX/Linux, data management, an introduction to MATLAB, and an introduction to Condor.

You can see a description of these classes and register for them by viewing our training calendar at http://www.itap.purdue.edu/training/calendar or browsing through the RCAC course listing at http://www.itap.purdue.edu/training/courselisting.cfm?ID=88. A complete listing of all ITaP course offerings may be found at http://www.itap.purdue.edu/training/.

Feel free to contact the Rosen Center training staff at rcac-training@purdue.edu if you have any questions.

Dr. Xiaoge Wang to speak on High Performance Computing in China --June 29th

Fri, 14 Jul 2006

In this talk, the current development and status of high performance computing in Mainland China will be introduced. An overview of two projects, ChinaGrid and CNGrid, which are the largest development of China�s infrastructure for high performance computing, will be presented, along with a brief introduction of CERNET, the network that serves as the backbone of ChinaGrid. Then, a closer look at the high performance computing related teaching and research activities in Tsinghua University (one of the top universities in science and engineering in mainland China) will be presented. A summary of some collaborative projects with other institutions, especially those with institutions outside China, are also presented. Altogether, a picture of the current state of high performance computing at Tsinghua University and its potentials for further collaborations with Purdue University is drawn for the audience.

Xiaoge Wang is a professor in the Department of Computer Science of Tsinghua University. She currently holds the position as the Associate Director of the Center for Operating System and Middleware R&D. Xiaoge Wang obtained her B.S and M.S degree in computer science from Tsinghua University in 1982 and 1984, respectively, and her PhD in computer science from UIUC in 1994. Her research interests include software infrastructure and software component technology for grid computing, pervasive computing, parallel and distributed computing, as well as the development of numerical methods and software of solving the problems arisen from the mathematical model of applications.

Date: June 29th

Time: 1:30-2:30pm

Place: RAWL 2058

Dr. Eng Lim Goh, Senior VP and CTO of SGI to speak

Fri, 14 Jul 2006

Dr. Eng Lim Goh, Senior Vice President and Chief Technology Officer of SGI will be on the Purdue campus on June 8th to make a scientific presentation. Dr. Goh will be in RAWL 2058 from 1:00pm -2:30pm speaking about the accelerated knowledge discovery from large data sets and about application profiling and its influence on performance.

Dr. Eng Lim Goh has been with SGI for 15 years, becoming one of the chief scientists in 1998 and chief technology officer in 2001. His tenure includes work in computer graphics algorithms and high performance computing (HPC) architectures.

In HPC, he oversees Project Ultraviolet, the goal of which is to design and build the company's next generation science-driven computer architecture. He is also the coauthor of SGI's recommendation to the high-end computing revitalization task force (HECRTF) for federal funding of key corresponding technologies. This proposal was reviewed by HECRTF in 2003 and judged to be one of the top submitted papers.

Dr. Goh is known as a proponent of next-generation computer systems designed specifically for customer applications performance. To this, he advocates computational density and a balanced multi-paradigm approach, across a globally addressable memory, to architectural design.

In computer graphics, Dr. Goh's current research interest is in the relationships between human visual perception and visual computing. He has been awarded two U.S. patents in this field. He is also leading a small research effort to investigate application-transparent, massively parallel advanced rendering.

In 2005, the IDG publication, InfoWorld, named Dr. Goh one of World's 25 most influential CTOs. He has also been named in the HPCwire list of "15 People to Watch."

Before joining SGI, Dr. Goh worked for Intergraph Systems, Schlumberger Wireline Netherlands, and Shell Research U.K. A Shell Cambridge University Scholar, he completed his Ph.D. research and dissertation on parallel architectures and computer graphics. He also holds a first-class honors degree in mechanical engineering from Birmingham University, U.K.

Register Now!! -- TeraGrid '06

Fri, 14 Jul 2006

The registration is now open for TeraGrid '06, Advancing Scientific Discovery. TeraGrid '06 is a forum for individuals and institutions who are interested in building and using cyberinfrastructure. The list of keynote and plenary speakers is impressive:

Dr. Arden Bement, Director of the National Science Foundation
Dr. Daniel Atkins, Director (Designate) of the Office of Cyberinfrastructure
Dr. Kelvin Droegemeier, Associate Vice President for Research, University of Oklahoma
Professor Donna Cox, Director of Visualization and Experimental Technologies, NCSA

Important Dates

5/11 Deadline for conference hotel room rate reservations
5/26 Deadline for early registration fees

Complete conference information available at the site listed below.

Interplay: Dancing on the Banks of Packet Creek

Fri, 14 Jul 2006

Real time multi-faceted telematic event consisting of six simultaneous performances occurring in six states throughout North America. Performers including actors, dancers, and musicians will be located at six different sites. The performers will also interact with computer graphics and audio. Each site will be creating its own artistic performance – with the six performances all coming together via the Access Grid to create a single integrated performance. Audience members will be able to see and hear the activities taking place at all of the remote sites in addition to the action at their own local site.

Participants: Purdue University Envision Center for Data Perceptualization, Alaska Fairbanks Arctic Region Supercomputing Center, University of Utah Center for High Performance Computing, Boston University, University of Maryland, Ryerson University. Toronto, Canada

Sponsored by the Purdue University Envision Center for Data Perceptualizaton and Patti and Rusty Rueff Department of Visual and Performing Arts.

March 31-April 2, 2006

Friday and Saturday, 9:00pm

Sunday, 6:00pm

All seating free but must be reserved in advance: Contact Purdue Envision Center 496-7888.

Protein Identification in Mass Spectrometry Based Proteomics Workshop

Fri, 14 Jul 2006

A regional educational workshop offered collaboratively by the Bindley Bioscience Center and the Rosen Center for Advanced Computing (RCAC) will be held on April 11 from 8:30am-4:00pm. Xiang Zhang, Lead Scientist for Computational Life Sciences and Informatics will be the lecturer.

Participants in this 1-day workshop will learn to:

· Understand the principle of protein identification from tandem mass spectra.

· Comprehend the difference between database searching and de novo sequencing.

· Experience analyzing tandem mass spectra using existing informatics tools.

· Identify the importance of validating peptide identification results.

· Appreciate the complexity of peptide identification.

· Design searching strategies based on the biological sample.

· Recognize current capabilities available for protein identification at Bindley Bioscience Center.

Lecture, discussion, and computer laboratory experiences will be used. Database searching methods will be introduced in the morning session, with de novo sequencing methods introduced in the afternoon session. All material will be available on the web a week prior to the workshop. Graduate and postdoctoral level researchers in chemistry, biochemistry, biology, computer science, statistics, agriculture, pharmacy, and others are welcome. Registration is free but required; attendance is limited to 20 and will be reserved in the order that registration is received. Lunch will be provided.

To register: http://www.itap.purdue.edu/training/registration/?offeringid=890

New Date -- May 2 --Research Data and the Distributed Institutional Repository

Thu, 13 Jul 2006

Purdue Libraries is collaborating with ITaP on a project that will lead to further collaboration with the new Cyber Center at Discovery Park. The collaboration addresses issues involving the communication and exchange of research and solves problems in organizing, accessing, and sorting large-scale data and information.

Please join us for this presentation by D. Scott Brandt and Michael C. Witt on Tuesday, May 2nd, 1:30-3:00pm in the PMU Anniversary Drawing Room.

NSF Director, Arden Bement, to speak

Thu, 13 Jul 2006

NSF Director, Dr. Arden Bement, will be the conference keynote speaker for the first annual TeraGrid Conference -- TeraGrid '06, Advancing Scientific Discovery -- to be held in Indianapolis, Indiana, June 12-15. TeraGrid '06 is hosted by the Rosen Center for Advanced Computing of Purdue University and Indiana University at the IUPUI Conference Center.

TeraGrid '06 is calling for presentations, demonstrations, posters, and birds-of-a-feather. All interested individuals and organizations are invited to participate. Attendees will include scientists, reasearchers, faculty, postdocs, graduate and undergraduate students, high school teachers and students, representatives from federal agencies, business and industry representatives involved in grid computing products and services, and TeraGrid resource providers. The Conference is also calling for student entries in two competitions, the Tera Grid CI-Impact Student Contest and the TeraGrid Student Research Competition.

For more information on the conference or to submit an entry, please go to http://www.teragrid.org/events/2006conference/.

Would you like to experience virtual reality??

Thu, 13 Jul 2006

Would you like to experience virtual reality???

We are seeking subjects to participate in an experiment on virtual reality taking place at the Envision Center at Purdue University. The study would require about an hour of your time and includes an experience with the Envision FLEX virtual reality system. If you are over 18 years of age and interested in participating, please contact Dr. Laura Arns, 494-6432, arns@purdue.edu.

Blue Gene Consortium Day at IBM's Watson Research Center

Thu, 13 Jul 2006

Blue Gene Watson (BGW) Consortium Day provides an opportunity for members of the Consortium to run applications or computer science projects on IBM's 20 rack, 100 Tflop BG system.

The next opportunity is March 29-30, 2006. Information can be found on the link listed below.

This is a reminder that calls for nominations must be submitted no later than 2/17/06. Forms are available below.

BGW Application Nomination Form

BGW Consortium CS Proposal Nomination Form