ITaP surveying faculty and staff about research computing needs and new supercomputer
March 12, 2010
ITaP is planning a new community cluster to be built this spring and is gathering information about what Purdue faculty and staff may need either as a part of the new cluster or in computers for departmental or individual use. The combined purchase allows ITaP to negotiate a better price with vendors in both cases and offers participants more for their money.
An online survey covers some configuration options, but others are possible and the list may change after ITaP gets feedback from faculty and staff and new product information from vendors. There is space in the survey form to make suggestions about other configurations or to ask questions. The survey is available at:
Information about the community cluster program also is available by emailing:
The deadline for completing the survey is April 2. After ITaP receives input from potential users and vendors, ITaP staff will make a list of configurations to send to the vendors for bidding. ITaP sends updates about the community cluster through the community cluster mailing list. To subscribe go to:
ITaP built community clusters, named Steele and Coates, in 2008 and 2009, which increased the supercomputing power available to Purdue researchers from just 14 teraflops in 2006 to more than 160 in 2009. That ranks Purdue near the top nationwide and first in the Big Ten among academic institutions not dependent on a national laboratory or a national supercomputing center.
In 2009, 58 researchers from 22 Purdue departments and seven schools bought into the Coates cluster. The cluster is being used in diverse ways, from modeling climate change and designing new medicines to studying Wikipedia’s social structure and simulating next-generation nanoscale electronics.
Through community clustering, Purdue’s Rosen Center for Advanced Computing, the research and discovery arm of ITaP, pools funds to make more computing power available than faculty and campus units could afford individually. ITaP installs, administers and maintains those systems, including security, so researchers can concentrate on doing research not running a high-performance computing system.
A community cluster also maximizes the use of resources by sharing computing power researchers need only part of the time with their peers, who tap it during what would otherwise be idle time. Participants always have access to the computing power they purchase, and potentially more when they need it.
Writer: Greg Kline, science and technology writer, Information Technology at Purdue (ITaP), 765-494-8167 (office), 765-426-8545 (mobile), email@example.com