Advertisement

NEC Vector Supercomputer: Its Present and Future

  • Shintaro MomoseEmail author

Abstract

SX-ACE launched in 2014 is a successor model of the SX-9 vector parallel computer, which pursues a much higher sustained performance particularly in memory-intensive scientific applications. The major concept of SX-ACE is the provision of a world top-level single core performance of 64 GFlop/s, as well as the world largest memory bandwidth per single core of 256 GB/s in its maximum with a high power efficiency. It is also designed to make available a user-friendly environment that can be combined with PC clusters in targeting a wide range of application areas. Experimental results demonstrate that SX-ACE can provide a much higher sustained performance and a power efficiency compared with modern supercomputers especially for memory-intensive applications. Moreover, NEC has a plan to release a follow-on system of SX-ACE. The new system is aimed at incorporating standard features and use environments of PC clusters while maintaining a high sustained performance realized with SX-ACE. It is expected to cover more extensive market areas, including emerging big data analyses, as well as conventional scientific and engineering applications.

References

  1. 1.
    Momose, S., Hagiwara, T., Isobe, Y., Takahara, H.: The brand-new vector supercomputer, SX-ACE. In: Proceedings of International Supercomputing Conference (ISC 14), pp. 199–214 (2014)Google Scholar
  2. 2.
    Momose, S.: SX-ACE processor: NECs brand-new vector processor. In: Proceedings of HOTCHIPS26, pp. 1.1.1–1.1.14 (2014)Google Scholar
  3. 3.
    Egawa, R., Momose, S., Komatsu, K., Isobe, Y., Musa, A., Takizawa, H., Kobayashi, H.: Early evaluation of the SX-ACE processor. In: SC14 Poster (International Conference on High Performance Computing, Networking, Storage and Analysis) (2014)Google Scholar
  4. 4.
    Momose, S.: SX-ACE, brand-new vector supercomputer for higher sustained performance I. In: Proceedings of the Joint Workshop on Sustained Simulation Performance, University of Stuttgart (HLRS) and Tohoku University, pp. 57–67 (2014)Google Scholar
  5. 5.
    Hoshi, N., Momose, S.: SX-ACE, the brand-new vector supercomputer for higher sustained performance II. In: Proceedings of the Joint Workshop on Sustained Simulation Performance, University of Stuttgart (HLRS) and Tohoku University, pp. 69–79 (2014)Google Scholar
  6. 6.
    Satoshi, N., Satoru, T., Norihito, N., Takayuki, W., Akihiro, S.: Hardware technology of the SX-9 (1) main system. NEC Tech. J. 3(4), 15–18 (2008)Google Scholar
  7. 7.
    Takahara, H.: NEC SX series vector supercomputer. In: Encyclopedia of Parallel Computing, vol. 4, pp. 1268–1277. Springer, Berlin (2011)Google Scholar
  8. 8.
    Soga, T., Musa, A., Shimomura, Y., Itakura, K., Okabe, K., Egawa, R., Takizawa, H., Kobayashi, H.: Performance evaluation on NEC SX-9 using real science and engineering applications. In: Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis, pp. 1–12 (2009)Google Scholar
  9. 9.
    Zeiser, T., Hager, G., Wellein, G.: The worlds fastest CPU and SMP node: some performance results from the NEC SX-9. In: Proceedings of IEEE International Symposium on Parallel & Distributed Processing (IPDPS2009), pp. 1–8 (2009)Google Scholar
  10. 10.
    McCalpin, J.: Memory bandwidth and machine balance in current high performance computers. In: IEEE Computer Society Technology Committee on Computer Architecture (TCCA) Newsletter, pp. 19–25 (1995)Google Scholar
  11. 11.
    Kerbyson, D.J., Barker, K.J., Vishnu, A., Hoisie, A.: Comparing the performance of Blue Gene/Q with leading Cray XE6 and InfiniBand systems. In: Proceedings of 2012 IEEE 18th International Conference on Parallel and Distributed System, pp. 556–563 (2012)Google Scholar
  12. 12.
    The Himeno benchmark, http://accc.riken.jp/2444.htm
  13. 13.
    Sato, Y., Inoguchi, Y., Luk, W., Nakamura, T.: Evaluating reconfigurable dataflow computing using the himeno benchmark. In: Proceedings of International Conference on Reconfigurable Computing and FPGAs (ReConFig), pp. 1–7 (2012)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2015

Authors and Affiliations

  1. 1.IT Platform DivisionNEC CorporationTokyoJapan

Personalised recommendations