HP retains first position in TOP500 supercomputer number of systems ranking

Palo Alto 15 November 2002 HP again ranks as the no. 1 supercomputing supplier on the TOP500 Supercomputer Sites list. Led by the HP Superdome server and AlphaServer systems, HP has for the second consecutive time more sites in the TOP500 list than any other technology provider.

HP posted 137 entries on the list, more than 25 percent of the sites, with HP Superdome servers notching 112 entries and HP AlphaServer systems registering 23 and leading the very high end. Also, for the first time, installations using HP's new Itanium 2-based systems appeared on the list and made it into the top 100.

HP had the most sites in the top 10 with four. This includes the ASCI Q supercomputer at Los Alamos National Laboratory, whose two partitions rank as the most powerful supercomputers in the United States.

HP Superdome systems ranked first in the enterprise category, a category of servers used by companies in both the commercial and technical markets, including the telecommunications, financial services, automotive, and aerospace industries.

"With HP Superdome servers providing industry-leading enterprise computing capabilities, HP AlphaServer systems delivering the best performance for advanced science and design, and the new HP Itanium 2-based systems delivering outstanding performance and scalability, HP offers the best and broadest portfolio of adaptive infrastructure solutions for the high performance computing market", stated Mr. Prather.

HP supercomputer installations running in some of the world's premier technical computing environments ranked on the TOP500 list include the following sites.

Los Alamos National Laboratory

The HP AlphaServer SC system at the National Nuclear Security Administration's Los Alamos National Laboratory in New Mexico ranks as the two most powerful supercomputers in the United States and numbers two and three on the worldwide list with its partitioned systems.

In August 2000, the NNSA selected HP to build a 30 TeraOPS (trillions of operations per second) system using more than 12,000 Alpha processors. The system is currently configured in two 10 TeraOPS clusters, composed of 1024 ES45 systems each. These segments will be joined together early next year into one system.

Code named "Q", the supercomputer is the latest in the Advanced Simulation and Computing programme within NNSA's Stockpile Stewardship Programme. That programme uses modelling and simulation for surveillance of the U.S. nuclear stockpile, experiments, non-nuclear tests, archived data, and fundamental science to assess and certify the safety, security, and reliability of nuclear weapons without underground nuclear testing.

Pittsburgh Supercomputing Center

Coming in sixth on the list was the Terascale Computing System, an HP AlphaServer SC45 supercomputer with more than 3000 processors, installed at the Pittsburgh Supercomputing Center. This system, which provides computing capability for use by U.S. researchers in all science and engineering disciplines, is the most powerful supercomputer in the United States committed to public research.

The National Science Foundation recently selected the Pittsburgh Supercomputing Center and its two sister supercomputing centres to create new technological infrastructure that will harness the nation's most powerful computing systems for open research into a seamless "grid". The TeraGrid will be the first wide-area computational grid encompassing terascale systems of differing architectures.

French Atomic Commission

Ranking seventh on the list is the "Tera" supercomputer deployed at the French Atomic Commission. The AlphaServer (3.98 TFLOP/s) system, which is Europe's most powerful supercomputer, is being used to perform digital simulation and testing of nuclear weapons. In just one second, the supercomputer can do the work of 30,000 mathematicians working night and day for five years on their handheld calculators.

Pacific Northwest National Laboratory

The U.S. Department of Energy's (DOE) Pacific Northwest National Laboratory (PNNL) prototype of its Itanium 2 processor-based supercomputer came in at 61 on the list.

When the installation is complete in late 2003, PNNL's HP supercomputer with more than 1400 Itanium 2 processors is expected to achieve more than 11 teraflops performance and will be both the most powerful Linux system, and, when connected to the DOE Science Grid, the largest supercomputer attached to a computer grid anywhere in the world. The PNNL supercomputer will allow researchers to apply computational science to address key scientific challenges.

Ad Emmen

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