Purdue's new research supercomputer designed to fit a variety of computational needs

February 10, 2015

Purdue’s latest Community Cluster Program research supercomputer offers faculty and campus units more than just one big community in which to settle their research computations. The new community cluster has neighborhoods, too.

Need big memory? There’s a subdivision for you. Need a place for high-throughput computing? Got it. And don’t worry if you need high-performance parallel computation. There’s plenty of room for that, too, just as in past community clusters.

Purdue researchers can find out more about the new cluster, to be built by ITaP Research Computing this spring, at a faculty luncheon set for noon to 1:30 p.m. Friday, Feb. 27, in the Lawson Computer Science Building, Room 1142.

The luncheon also is an opportunity to provide input on the new cluster. Researchers can email rcac-help@purdue.edu with questions or advice as well.

The new cluster will run on the same shared, community model that has given Purdue the best supercomputing infrastructure in the nation for use by researchers on a single campus. But subdividing it to address specific computational demands of different types of research will better serve more users, says Preston Smith, director of research services and support for ITaP Research Computing.

“If you need big memory or high throughput, this will be much better for you,” Smith says. “If you do high-performance parallel computing, we’ve got you covered as always.”

The new cluster is the latest research computing system offered to Purdue faculty through the Community Cluster Program, under which ITaP and faculty partners have enabled the building of six TOP500-class supercomputers since 2008. The new cluster will be the seventh.

Through community clustering, faculty partners and ITaP make more computing power available for Purdue research than faculty and campus units could individually afford. ITaP installs, administers and maintains the community clusters, including security, software installation and user support, so researchers can concentrate on doing research rather than on running a high-performance computing system.

Community clustering also maximizes use by sharing computing power among the faculty partners whenever it is idle. Researchers always have ready access to the capacity they purchase, and potentially to much more if needed.

In addition to computational and archival storage for the community clusters, ITaP also operates the Research Data Depot for faculty, whether cluster users or not. The Research Data Depot is available to any Purdue researcher or campus unit in need of a reliable, secure high-capacity central solution for storing large, active research data sets at a competitive price and easily sharing them among on- and off-campus collaborators.

For more information on the Research Data Depot or the community clusters, contact Preston Smith, psmith@purdue.edu or 49-49729.

Originally posted: February 10, 2015  10:44am