National Science Foundation taps into IBM-Google computer cluster

The National Science Foundation announced this week that it had reached an agreement with Google and IBM that would let the academic research community conduct experiments through the companies'1,600-processor computer cluster.

Google, IBM and the NSF's Computer Science and Engineering Directorate will be launching a joint initiative called the Cluster Exploratory (CluE) that will grant the academic research community access to the Google-IBM academic cluster. The NSF  also says that the cluster will give researchers access to resources that would otherwise have otherwise been prohibitively expensive.

Willy Chiu, the vice president software strategy for IBM, says that the new program will "accelerate research on Internet-scale computing and drive innovation to fuel the applications of the future." The NSF says its Computer Science and Engineering Directorate will solicit research proposals from academics from around the country and decide which ones would best utilize the cluster's capabilities. While Google and IBM will cover the costs of operating and maintaining the cluster for the researchers' work, the NSF says it will not pay either of the companies for these activities.

Google and IBM first announced last fall that they were using the cluster as part of a joint initiative to help computer science students gain more knowledge of highly parallel-computing practices. Parallel computing is a method for computers to more quickly carry out large-scale tasks by simultaneously handling several different instructions through multiple processors.

Researchers at the University of Maryland, for instance, developed a parallel processing desktop computer this summer that they say runs 100 times faster than today's PCs.

Google says it has been "encouraged" so far by the results of the parallel computing initiative that launched at several U.S. universities as part of the Academic Cluster Computing Initiative, and that the NSF was a "natural partner" to help expand the cluster's reach. The University of Washington was the first university to join the initiative, with Carnegie Mellon University, the Massachusetts Institute of Technology, Stanford University , the University of California-Berkeley and the University of Maryland also piloting the program.