IBM delivers new Itanium-based systems, including Linux supercomputer

Armonk 31 May 2001 IBM introduced new Intel Itanium-based IBM eServer and IntelliStation workstations which will form a new Linux supercluster at the National Center for Supercomputing Applications (NCSA) at the University of Illinois at Urbana- Champaign. Capable of 1 Tflop/s, a cluster of 160 new IBM Itanium-based systems will be the most powerful Linux supercluster in academia, IBM claims.

Scientists will use NCSA's IBM eServer clusters to study a wide range of research problems, from how gas flows over surfaces at supersonic speeds to simulations of black hole collisions, which demonstrate principles of Einstein's General Relativity Theory. The new Itanium-based IBM eServer cluster will complement the cluster of 512 IBM eServer x330's installed earlier this year.

The new IBM eServer x380, IBM's first Itanium-based server, was designed as the industry standard reference platform for IA-64 development efforts. Available with up to four processors, the x380 delivers increased performance for data intensive applications such as data mining and data warehousing, online transaction processing and security applications. In addition, the x380 is available with up to 64GB(1) of memory permitting in-memory processing of data intensive applications, which helps avoid bottlenecks. System also includes hot swap and redundant fans and four power supplies, along with hot- swap PCI technology.

Targeted towards bringing high-end Intel computing into the technical environment and data center, the x380 can scale up to four processors and scale out to hundreds of nodes using industry standard clustering technology.

The new IntelliStation Z Pro announced today is IBM's first workstation to take advantage of Itanium, creating new solutions for engineers, media creators and scientists.

Working in collaboration with the University of Minnesota, NCSA scientists using the new IntelliStation Z Pro to study gas flow dynamics have achieved single-processor performance in excess of 650 megaflops. Engineers can take advantage of floating point performance to run compute-intensive applications, such as finite analysis, on lower-cost platforms. And media creators can design more life-like animations while reducing rendering time.


Ad Emmen

[News on Advanced IT][Calendar][Analysis][IT in Medicine]