Misplaced Pages

DEGIMA

Article snapshot taken from Wikipedia with creative commons attribution-sharealike license. Give it a read and then ask your questions in the chat. We can research this topic together.
(Redirected from DEGIMA (computer cluster)) High performance computer cluster

The DEGIMA (DEstination for Gpu Intensive MAchine) is a high performance computer cluster used for hierarchical N-body simulations at the Nagasaki Advanced Computing Center, Nagasaki University.

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 motherboard. 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.

The name of the system is also derived from the name of a small artificial island called "Dejima" in Nagasaki.

See also

References

  1. 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
  2. The Green500 June 2011 Environmentally Responsible Supercomputing, The Green500 List
  3. 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 Misplaced Pages by expanding it.

Categories: