The system consists of a 144-node cluster of PCs connected via an InfiniBand interconnect. Each node is composed of 2.66 GHz Intel Core i7 920 processor, two GeForce GTX295 graphics cards, 12 GB DDR3-1333 memory and Mellanox MHES14-XTC SDR InfiniBand host adapter on MSI X58 pro-E mother board. Each graphics card has two GT200 GPU chips. As a whole, the system has 144 CPUs and 576 GPUs. It runs astrophysical N-body simulations with over 3,000,000,000 particles using the Multiple-Walk parallel treecode. The system is noted for being highly cost and energy-efficient, having a peak performance of 111 TFLOPS with an energy efficiency of 1376 MFLOPS/watt. The overall cost of the hardware was approximately US$500,000.
- Hamada T. et al. (2009) A novel multiple-walk parallel algorithm for the Barnes–Hut treecode on GPUs – towards cost effective, high performance N-body simulation. Comput. Sci. Res. Development 24:21-31. doi:10.1007/s00450-009-0089-1
- The Green500 June 2011 Environmentally Responsible Supercomputing, The Green500 List
- Hamada T., Nitadori K. (2010) 190 TFlops astrophysical N-body simulation on a cluster of GPUs. In Proceedings of the 2010 ACM/IEEE International Conference for High Performance Computing, Networking, Storage and Analysis (SC '10). IEEE Computer Society, Washington, DC, USA, 1-9. doi:10.1109/SC.2010.1
|This supercomputer-related article is a stub. You can help Wikipedia by expanding it.|