Vendor Spotlight:

ITALIAN, GERMAN INSTITUTES ORDER CRAY XD1 SUPERCOMPUTERS

Cray Inc. announced that the National Institute of Nuclear Physics (INFN) is Italy's first disclosed customer for the new Cray XD1 supercomputer. Financial details were not disclosed.

A Cray XD1 supercomputer is scheduled to be installed later this year at INFN's Pisa section, where it will support research in theoretical and experimental physics.

"The Cray XD1 system's innovative design, especially the interconnect architecture, will allow us to deliver affordable high performance computing to the physicists' desktops and will help us reach new frontiers in computing for physics research," said Professor Paolo Rossi, chair of the Pisa Physics Department.

According to Professor Rino Castaldi, director of INFN's Pisa section, "The Cray XD1 supercomputer's use of AMD Opteron processors was another advantage. Thanks to our strong collaboration with AMD, we were among the first to adopt AMD Opteron processors in our Grid computing environment. The AMD Opteron has rapidly become the preferred processor in our Computing Center, both for theoretical physics calculations and experimental data analysis."

"We are excited to form this new relationship with INFN, one of the world's leading institutions for fundamental physics research. The Cray XD1 platform is purpose-built to meet HPC challenges and is an attractive alternative to clusters assembled from general-purpose business servers," said Ulla Thiel, vice president of Cray Europe. "Customers in many disciplines around the world are now turning to Cray for this high performance solution."

KONRAD ZUSE CENTER FOR INFORMATION TECHNOLOGY IN BERLIN ORDERS CRAY XD1

Cray Inc. also announced that it has received an order for a Cray XD1 supercomputer from Germany's renowned Zuse Institute Berlin (ZIB). Financial details were not disclosed.

ZIB will use the Cray XD1 system primarily for computational chemistry applications. ZIB will also exploit the system's unique field programmable gate array (FPGA) capabilities to advance the use of reconfigurable computing to accelerate life sciences applications.

"We are expanding our long-term relationship with Cray because we are excited about the Cray XD1 supercomputer's combination of exceptional performance and new innovative features," said Alexander Reinefeld, head of Computer Science. "This system has strong potential for advancing the work of some of our most demanding users."

"We are excited to deepen our partnership again with ZIB, one of the leading institutions for scientific computing. The Cray XD1 platform is purpose- built to meet HPC challenges and is an attractive alternative to clusters assembled from general-purpose business servers," said Ulla Thiel, vice president of Cray Europe. "Customers in many disciplines around the world are now turning to Cray for this high performance solution."

About the Cray XD1 Supercomputer

The Cray XD1 supercomputer features the direct connect processor (DCP) architecture, which removes PCI bottlenecks and memory contention to deliver superior sustained performance. The Cray XD1 has the lowest latency of any current HPC system, with MPI latency of 1.8 microseconds in measured tests conducted using the Ohio State MPI benchmark. The tests show that the Cray XD1 ships messages with four times lower MPI latency than common cluster interconnects such as Infiniband, Quadrics or Myrinet, and 30 times lower than Gigabit Ethernet employed in lowest-cost clusters. The Cray XD1's interconnect delivers twice the throughput of Infiniband for messages up to 1 KB and 60 percent higher throughput for very large messages.

The Linux/Opteron system runs x86 32 and 64-bit codes. Field programmable gate arrays (FPGAs) are available to accelerate applications, and the Active Manager subsystem provides single system command and control and high availability features. A 3VU (5.25") chassis provides 12 compute processors, 58 peak gigaflops, 96 GB/second aggregate switching capacity, 1.8-microsecond MPI interprocessor latency, 84 GB maximum memory and 1.5 TB maximum disk storage. A 12-chassis rack provides 144 compute processors, up to 691 peak gigaflops, 1TB/second aggregate switching capacity, 2 microsecond MPI interprocessor latency, 922 GB/second aggregate memory bandwidth, 1 TB maximum memory and 18 TB maximum disk storage.

About Cray Inc.

As the global leader in high performance computing (HPC), Cray provides innovative supercomputing systems that enable scientists and engineers in government, industry and academia to meet both existing and future computational challenges. Building on years of experience in designing, developing, marketing and servicing the world's most advanced supercomputers, Cray offers a comprehensive portfolio of HPC systems that deliver unrivaled sustained performance on a wide range of applications. Go to http://www.cray.com for more information.