The TOP500 project ranks and details the 500 most powerful publicly-known computer systems in the world. The project was started in 1993 and publishes an updated list of the supercomputers twice a year. The project aims to provide a reliable basis for tracking and detecting trends in high-performance computing and bases rankings on HPL, a portable implementation of the High-Performance LINPACK benchmark for distributed-memory computers. The TOP500 list is compiled by Hans Meuer of the University of Mannheim, Germany, Jack Dongarra of the University of Tennessee, Knoxville, Erich Strohmaier and Horst Simon of NERSC/Lawrence Berkeley National Laboratory. The list is updated twice a year. The first of these updates always coincides with the International Supercomputer Conference in June, the second one is presented in November at the IEEE Super Computer Conference in the USA.
Contents |
Project history
In the early nineties, a new definition of supercomputer was needed to produce meaningful statistics. After experimenting with metrics based on processor count in 1992, the idea was born at the University of Mannheim to use a detailed listing of installed systems as the basis. Early 1993 Jack Dongarra was convinced to join the project with his Linpack benchmark. A first test version was produced in May 1993, partially based on data available on the Internet, including the following sources:[1][2]
- Mannheim Supercomputer Statistics 1986-1992
- one was "List of the World's Most Powerful Computing Sites" maintained by Gunter Ahrendt
- another was through David Kahaner, who had an immense amount of data.
The information from those sources was used for the first two lists. Since June 1993 the TOP500 is produced bi-annually based on site and vendor submissions only.
The systems ranked #1 since 1993
- IBM Blue Gene/L (since 2004.11)
- NEC Earth Simulator (2002.06 - 2004.11)
- IBM ASCI White (2000.11 - 2002.06)
- Intel ASCI Red (1997.06 - 2000.11)
- Hitachi CP-PACS (1996.11 - 1997.06)
- Hitachi SR2201 (1996.06 - 1996.11)
- Fujitsu Numerical Wind Tunnel (1994.11 - 1996.06)
- Intel Paragon XP/S140 (1994.06 - 1994.11)
- Fujitsu Numerical Wind Tunnel (1993.11 - 1994.06)
- TMC CM-5 (1993.06 - 1993.11)
List as of November 2007
The following table gives the Top 10 positions of the 30th TOP500 List released during the SC07 conference, November 12, 2007 in Reno, Nevada.[3]
| Rank | Site Country Year installed |
Name Computer Processors |
Maker | Rmax Rpeak (Tflops) |
|---|---|---|---|---|
| 1 | Lawrence Livermore National Laboratory United States 2005 |
BlueGene/L eServer Blue Gene Solution 212992 (Power) |
IBM | 478.2 596.4 |
| 2 | Jülich Research Centre Germany 2007 |
JUGENE BlueGene/P Solution 65536 (Power) |
IBM | 167.3 222.8 |
| 3 | New Mexico Computing Applications Center United States 2007 |
SGI Altix ICE 8200 14336 (Xeon), InfiniBand |
SGI | 126.9 172.0 |
| 4 | Computational Research Laboratories India 2007 |
EKA Cluster Platform 3000 14240 (Xeon), InfiniBand |
HP | 117.9 170.9 |
| 5 | Swedish National Defence Radio Establishment Sweden 2007 |
Cluster Platform 3000 13728 (Xeon), InfiniBand |
HP | 102.8 146.4 |
| 6 | Sandia National Laboratories United States 2006 |
Red Storm Cray XT3 26569 (Opteron) |
Cray | 102.2 127.5 |
| 7 | Oak Ridge National Laboratory United States 2006 |
Jaguar Cray XT4/XT3 23016 (Opteron) |
Cray | 101.7 119.4 |
| 8 | IBM Thomas J. Watson Research Center United States 2005 |
BGW eServer Blue Gene Solution 40960 (Power) |
IBM | 91.3 114.7 |
| 9 | National Energy Research Scientific Computing Center United States 2007 |
Franklin Cray XT4/XT3 19320 (Opteron) |
Cray | 85.4 100.5 |
| 10 | Stony Brook University/Brookhaven National Laboratory United States 2007 |
New York Blue eServer Blue Gene Solution 36864 (Power) |
IBM | 82.2 103.2 |
Trends
A paper[4] made available in early 2007 presents the results of some statistical analysis of the Top 500 from 1993 to 2006. Some of the conclusions are:
- Each list exhibits a Zipf distribution
- Speeds increase by about 90% per year.
- The relative power of the bottom of the list as compared to the top of the list has increased each year.
Systems of interest
The MDGRAPE-3 supercomputer reportedly reached a one petaflops calculation speed, faster than any of those on the TOP500 list, though it does not qualify as a general-purpose supercomputer, and cannot run the benchmarking software used to gauge the speeds computed for the list. Grid computing systems are also not included on the list.
References
- ^ http://www.hpcwire.com/archives/10219.html
- ^ http://www.netlib.org/benchmark/top500/reports/report93/section2_16_4.html
- ^ http://sc07.supercomputing.org/
- ^ A note on the Zipf distribution of Top500 supercomputers, Matei Ripeanu, in IEEE Distributed Systems Online
