Recommended
More Related Content
Viewers also liked
Viewers also liked (20)
More from top500
More from top500 (11)
Recently uploaded
Recently uploaded (20)
June 2011 TOP500 Presentation
- 1. Highlights of the 37th TOP500 List ISC’11, Hamburg, Germany
- 2. 37th List: The TOP10
- 4. Annual Performance Increase of theTOP500
- 5. Performance Development 58.9PFlop/s 100 Pflop/s 8.16PFlop/s 10 Pflop/s 1 Pflop/s 40.2TFlop/s 100 Tflop/s SUM 10 Tflop/s N=1 1 Tflop/s 1.17 TFlop/s 100 Gflop/s 59.7 GFlop/s N=500 10 Gflop/s 1 Gflop/s 400 MFlop/s 100 Mflop/s
- 6. Projected Performance Development 1Eflop/s 100 Pflop/s 10 Pflop/s SUM 1 Pflop/s 100 Tflop/s N=1 10 Tflop/s 1 Tflop/s N=500 100 Gflop/s 10 Gflop/s 1 Gflop/s 100 Mflop/s
- 12. Vendors / System Share
- 13. Vendors (TOP50) / System Share
- 14. IBM Cray HP Fujitsu SGI
- 15. IBM Cray HP Fujitsu SGI
- 16. IBM Cray HP Fujitsu SGI
- 17. Countries / System Share
- 18. Asian Countries
- 21. IBM Cray HP Fujitsu SGI
- 22. intel Power SPARC AMD
- 25. Cores per Socket
- 26. Cores per Socket
- 27. Accelerators
- 34. Power Efficiency
- 35. Most Power Efficient Systems [Mflops/Watt]
- 36. Darker Colors are more Power efficient
- 37. BlueGene/Q Power Efficiency Cell Kcomputer
Editor's Notes
- (AICS) missing in name of first entryPleiades were #11 last timeSystem configuration: - 672 computer cabinets, 68544 nodes - 1 node consists of 1 CPU - 8 cores/CPU - 6-D Mesh/Torus Interconnect SPARC64 VIIIfx CPU: - 8 cores/CPU - 128 GFlops @ 2GHz
- Annual versus Moore’s Law1.85 versus 1.59
- Fujitsu, Dell, and Dawning have 2 each
- Satoshi just announced 2.4 PF/s system with nVidia FermiOct 2010 HP + nVidia + new features
- Satoshi just announced 2.4 PF/s system with nVidia FermiOct 2010 HP + nVidia + new features
- Satoshi just announced 2.4 PF/s system with nVidia FermiOct 2010 HP + nVidia + new features
- #50 is ASCI Purple
- Larger systems are more efficientInnovationin HPC is “top-down”
- K computer: "K" here draws upon the Japanese word "Kei" for 1016(ten quadrillions), representing the system's performance goal of 10 petaflops. In its original sense, "Kei" expresses a large gateway in Japanese, and it is hoped that the system will be a new gateway to computational science.