National Science Foundation Awards Pittsburgh Supercomputing Center $2.8M Toward the Purchase of World's Largest Coherent Shared-Memory System

PITTSBURGH, PA., July 29, 2010 - The National Science Foundation (NSF) has partially funded the Pittsburgh Supercomputing Center (PSC) to acquire a system that features SGI's (NASDAQ:SGI) newest scalable, shared-memory computing system and associated disks. The SGI® Altix® UV system features extremely large, coherent shared-memory and opens a new computational capability for U.S. scientists and engineers.


The SGI® Altix® UV system

The Altix UV system will have 4,096 cores, in 512 eight-core Intel Xeon (Nehalem) processors, with 32 terabytes of memory, organized as two connected 16-terabyte coherent shared-memory systems, making these the largest coherent shared-memory systems in the world. Coherence, a feature related to the synchrony of read-write operations by different processors within the system, is an important feature in many large data-analysis tasks.

PSC will integrate the new system into the TeraGrid, the NSF program of comprehensive cyberinfrastructure, greatly increasing the capability available for U.S. science and engineering research.

"Because of the extraordinary memory size and relative ease of programming made possible by the Altix UV shared-memory structure, scientists and engineers will be able to solve problems that were heretofore intractable," said PSC scientific directors Michael Levine and Ralph Roskies in a joint statement. "For many research communities, including data-analysis and many areas of computer science, it will open the door to use of high-performance computation and thereby expand the abilities of scientists to ask and answer questions."

"PSC has a long-standing history of breakthrough scientific research through technology leadership and industry collaboration," said Mark Barrenechea, SGI CEO. "SGI is excited to partner with PSC, NSF and the TeraGrid alliance to enable the next generation of research with SGI's Altix UV shared memory super computer."

In computer terms, "shared memory" means that a system's memory can be directly accessed from all of its processors, as opposed to distributed memory (in which each processor's memory is directly accessed only by that processor). Because all processors share a single view of data, a shared memory system is relatively easy to program and use.

Because of its shared-memory design, the new PSC system will complement other NSF systems, most of which are based on distributed-memory architectures.

The 4,096 processor cores and 32 terabytes of shared memory are interconnected using SGI's next-generation high-bandwidth, low-latency NUMAlink® 5 interconnect. This interconnect has specialized features that enable scalable shared-memory or message-passing applications to run with higher levels of parallel efficiency so that researchers can assign more processor cores simultaneously to the same task. This allows researchers to address larger problems and solve them more quickly.

Delivery of the new system is scheduled for the summer of 2010. Production use will begin in TeraGrid's Oct 2010 allocation cycle.

System status

  Bridges status

Bridges is up and running normally.

Featured Projects

All about Bridges

Bridges 4c stacked

Allocation requests now being accepted.


Data Exacell (DXC)


The Data Exacell (DXC) is a research pilot project to create, deploy, and test software and hardware building blocks to enable data analytics in scientific research. Learn More

XSEDE Service Provider

image gallery

PSC is a service provide of the Extreme Science and Engineering Discovery Environment (XSEDE), the most advanced, powerful, and robust collection of integrated advanced digital resources and services in the world.