The Green500 List - November 2010
The Green500 continues to strive towards the goal of raising awareness in the energy efficiency of supercomputing, and in turn, drive energy efficiency as a first-order design constraint - one that is on par with performance (or speed). In this edition, we have the serendipity of having four uniquely green supercomputers at the top of the Green500:
- IBM BlueGene/Q, the "Greenest Supercomputer in the World"
- GRAPE-DR, the "Greenest Exotic Supercomputer in the World"
- Tsubame 2.0, the "Greenest Production Supercomputer in the World"
- EcoG, the "Greenest Self-Built Supercomputer in the World"
The IBM Blue Gene/Q prototype supercomputer is the third installment of a series of energy-efficient Blue Gene supercomputers, following in the footsteps of Blue Gene/L and Blue Gene/P. (In 2004, Blue Gene/L was the supercomputer that supplanted the Japanese Earth Simulator, which in turn, had created a "Computenik" event in 2002 when it shattered U.S. supercomputer hegemony.) While IBM BlueGene/Q prototype supercomputer sits at the top of the heap, accelerator-based supercomputers continue to dominate the remaining top slots of the Green500, holding 8 of the top 10 slots. Accelerators refer to the use of dedicated hardware to perform computations faster than a traditional processor, also known as a central processing unit (CPU). Given that the list, which is released twice a year - once in June and once in November, was similarly dominated by accelerator-based supercomputers in June, one might argue that this year’s Green500 was "The Year of the Accelerator."
For the measured systems on the Green500, the accelerator-based supercomputers on The Green500 List produce an average efficiency of 756 MFLOPS/W whereas the other measured supercomputers on the list only produce an average efficiency of 211 MFLOPS/W. That makes the accelerator-based supercomputers on the Green500 nearly 3.5 times more energy efficient than their non-accelerated counterparts on the list.
Listed below are the The Green500's Top 10 most energy-efficient supercomputers in the world as of November 2010.
| Green500 Rank | MFLOPS/W | Site* | Computer* | Total Power (kW) |
| 1 | 1684.20 | IBM Thomas J. Watson Research Center | NNSA/SC Blue Gene/Q Prototype | 38.80 |
| 2+ | 1448.03 | National Astronomical Observatory of Japan | GRAPE-DR accelerator Cluster, Infiniband | 24.59 |
| 2 | 958.35 | GSIC Center, Tokyo Institute of Technology | HP ProLiant SL390s G7 Xeon 6C X5670, Nvidia GPU, Linux/Windows | 1243.80 |
| 3 | 933.06 | NCSA | Hybrid Cluster Core i3 2.93Ghz Dual Core, NVIDIA C2050, Infiniband | 36.00 |
| 4 | 828.67 | RIKEN Advanced Institute for Computational Science | K computer, SPARC64 VIIIfx 2.0GHz, Tofu interconnect | 57.96 |
| 5 | 773.38 | Universitaet Wuppertal | QPACE SFB TR Cluster, PowerXCell 8i, 3.2 GHz, 3D-Torus | 57.54 |
| 5 | 773.38 | Universitaet Regensburg | QPACE SFB TR Cluster, PowerXCell 8i, 3.2 GHz, 3D-Torus | 57.54 |
| 5 | 773.38 | Forschungszentrum Juelich (FZJ) | QPACE SFB TR Cluster, PowerXCell 8i, 3.2 GHz, 3D-Torus | 57.54 |
| 8 | 740.78 | Universitaet Frankfurt | Supermicro Cluster, QC Opteron 2.1 GHz, ATI Radeon GPU, Infiniband | 385.00 |
| 9 | 677.12 | Georgia Institute of Technology | HP ProLiant SL390s G7 Xeon 6C X5660 2.8Ghz, nVidia Fermi, Infiniband QDR | 94.40 |
| 10 | 636.36 | National Institute for Environmental Studies | GOSAT Research Computation Facility, nvidia | 117.15 |
* Performance data obtained from publicly available sources including TOP500
