Within the top ten supercomputers, the National Center for Supercomputing Applications (NCSA), at the University of Illinois Urbana-Champaign has the most powerful open system interconnect cluster, named Abe. NCSA has chosen Cisco SFS 7000 series InfiniBand products for building a 90 Teraflop compute fabric, with over 1200 interconnected X86 servers. The Cisco InfiniBand interconnect provides high-speed communication capability, enabling users to run tightly coupled applications that achieve high levels of scaling. NCSA is one of the five original centres in the National Science Foundation's Supercomputer Centers Programme. The centre operates some of the world most powerful supercomputers, and is recognized as an international expert in deploying robust high-performance computing resources and in working with research communities to develop new computing and software technologies.

"Abe supports computationally challenging problems in production jobs typically using over 1000 cores", stated Mike Pflugmacher Assistant Director of Computing and Data Management at NCSA. "A stable, high performance interconnect is an absolute requirement for our needs, and we rely on Cisco's considerable expertise in implementing very large systems."

At the Statoil facilities in Stavanger, Norway, Cisco InfiniBand technology is used to interconnect the largest supercomputer in Scandinavia, ranked no. 151 in the top 500, and capable of executing twelve-thousand billion calculations (12 teraflops) per second with a cluster of 256 interconnected servers with 1,024 processors. The supercomputer is important for the realization of next generation oil exploration, helping to analyse many interpretations with different models to determine where the company should drill.

"Computing power plays a decisive role in providing better images of the geology several kilometers under the subsurface, especially in areas with demanding geological structures", stated Mr. Halsetrønning, project manager for the server cluster at Statoil. "We believe the combination of Cisco InfiniBand switches and HP servers provides a leading high performance clustering solution today to help resolve our complex models in a timely and accurate manner."

The Stanford University Bio-X2 cluster is the result of an NSF-funded research proposal submitted by 21 Bio-X affiliated faculty, representing 13 departments and 4 schools at Stanford. The purpose of the cluster is to facilitate biological research problems ranging in scale from molecules to organisms. This no. 54 ranked supercomputer runs 276 Dell PowerEdge 1950 servers using dual quad Intel Xeon processors. The servers are interconnected with InfiniBand Dual-Data Rate to a central Cisco SFS 7024D 288 port switch, and this fabric is the heart of the cluster, enabling multi-node parallel computations and the use of the Lustre file system for high performance data access.

"The Bio-X2 supercluster is used to solve complex problems and facilitate innovative interdisciplinary research at Stanford University", stated Tanya Raschke, IT Manager at the James H. Clark Center. "Consolidating compute resources enables researchers to solve much larger problems than would be possible if each department were funding and managing their own cluster. The high quality of the Cisco DDR Server Fabric Switch line allows us to focus less on managing the cluster and more on doing cutting-edge life sciences research."

One of the largest computing centers in the world, Cineca provides computing power for 65 Italian universities, research and government organizations. Cineca recently deployed Cisco high performance computing (HPC) InfiniBand and Ethernet networks to help researchers accelerate projects, make better investment decisions, and reduce research and development costs. "As one of Europe's leading computing centres, demand for our high-performance computing services and facilities to support vital research and development projects is increasing all the time", stated Sanzio Bassini, director of Cineca's System and Technology Department. "With our latest technology, we can now help with projects such as assessing and improving risk management if Mount Vesuvius erupts, or developing a new generation in molecular dynamics."