Skip to main content
Premium Trial:

Request an Annual Quote

UB s Bioinformatics Computer Debuts at #22 on Top500 List; Clusters Surge to Prominence

Premium

Bioinformatics has made its mark in the supercomputing world: The 4,000-processor Dell cluster installed at the Buffalo Center of Excellence in Bioinformatics in September made it to the #22 spot in the biannual Top500 list of the world’s fastest supercomputers.

Not only is this the highest that a dedicated life science computer has ever climbed on the list, but it exemplifies the quick rise of cluster computing through the high-performance ranks in recent years — a trend further borne out by the appearance of two PC-based clusters in the top ten of the current list (a 5.7-Tflop/s Linux Networx/ Quadrics cluster at Lawrence Livermore National Laboratory at #5 and a 3.3-Tflop/s HPTi/Myrinet cluster at the National Oceanic and Atmospheric Administration’s Forecast Systems Laboratory at #8).

The fastest cluster computer on the June list was an 825-Gflop/s AMD-based system that ranked #35 and now stands at #64.

The twentieth edition of the Top500 list, released last week at the Supercomputing Conference 2002 in Baltimore, Md., boasts a total of 55 Intel-based and 8 AMD-based PC clusters, up from 42 and 7, respectively, six months ago. Including Sun and Alpha-based clusters, the November Top500 list contains a total of 93 clusters, up from 80 in June.

The November list includes 14 cluster computers described as “self-made,” up from 12 six months ago and only 6 one year ago.

In addition to the Buffalo bioinformatics cluster, several other life science-related supercomputers made it to the current version of the list: IBM claims the #111 spot with a 536-Gflop/s system installed at an undisclosed pharmaceutical company; the #128 position with its 468-Gflop/s system at Applera and Celera Genomics; and also holds positions #253 and #353 with two systems for Bayer — one at 250 Gflop/s and one at 212 Gflop/s. SGI has two systems installed at Kyoto University’s Institute for Chemical Research, which are dedicated in part to genomics and bioinformatics research. The 406-Gflop/s and 210-Gflop/s systems give SGI the #155 and #362 spots, respectively. Finally, HP holds the #218 spot with a 296-Gflop/s system at Agilent.

Several life science firms who made it onto the last version of the Top500 list didn’t make the cut-off speed of 195.8 Gflop/s this time around (up from 134.3 Gflop/s in June): Syrrx, Inpharmatica, Aventis Pharma, and Amgen were all bumped from the current list.

Overall, NEC’s 35.9-Tflop Earth Simulator is still far and away the top machine — the next-fastest systems are identical 7.7-Tflop HP AlphaServer systems at Los Alamos National Lab. Tflop-scale systems are becoming commonplace, however. The number of systems exceeding the 1 Tflop/s mark on the Linpack benchmark is now 47, more than double the June list, which included 23 computers above 1 Tflop/s.

Also unchanged were the leading manufacturers by total performance and number of systems. IBM retained the number-one spot in total performance of all the installed systems, with 92.5 Tflop/s compared to HP’s 64.8 Tflop/s and NEC’s 42.7. HP, however, claims 137 total systems compared to IBM’s 129.

The Top500 list is compiled by Hans Meuer of the University of Mannheim, Germany; Erich Strohmaier and Horst Simon of Lawrence Berkeley National Laboratory; and Jack Dongarra of the University of Tennessee. The list ranks computers based on maximal performance acheived on the Linpack benchmark, rather than theoretical peak performance.

The complete list is available at www.top500.org.

— BT

File Attachments

Filed under

The Scan

And Back

The New York Times reports that missing SARS-CoV-2 genome sequences are back in a different database.

Lacks Family Hires Attorney

A lawyer for the family of Henrietta Lacks plans to seek compensation from pharmaceutical companies that have used her cancer cells in product development, the Baltimore Sun reports.

For the Unknown

The Associated Press reports that family members are calling on the US military to use new DNA analysis techniques to identify unknown sailors and Marines who were on the USS Arizona.

PLOS Papers on Congenital Heart Disease, COVID-19 Infection Host MicroRNAs, Multiple Malformation Mutations

In PLOS this week: new genes linked to congenital heart disease, microRNAs with altered expression in COVID-19, and more.