Advanced Computing and Optimization Infrastructure for Extremely Large-Scale Graphs on Post Peta-Scale Supercomputers

  • Katsuki Fujisawa
  • Toyotaro Suzumura
  • Hitoshi Sato
  • Koji Ueno
  • Yuichiro Yasui
  • Keita Iwabuchi
  • Toshio Endo
Conference paper
Part of the Mathematics for Industry book series (MFI, volume 13)

Abstract

In this paper, we present our ongoing research project. The objective of this project is to develop advanced computing and optimization infrastructures for extremely large-scale graphs on post peta-scale supercomputers. We explain our challenge to Graph 500 and Green Graph 500 benchmarks that are designed to measure the performance of a computer system for applications that require irregular memory and network access patterns. The 1st Graph500 list was released in November 2010. The Graph500 benchmark measures the performance of any supercomputer performing a BFS (Breadth-First Search) in terms of traversed edges per second (TEPS). We have implemented world’s first GPU-based BFS on the TSUBAME 2.0 supercomputer at Tokyo Institute of Technology in 2012. The Green Graph 500 list collects TEPS-per-watt metrics. In 2014, our project team was a winner of the 8th Graph500 benchmark and 3rd Green Graph 500 benchmark. We also present our parallel implementation for large-scale SDP (SemiDefinite Programming) problem. We solved the largest SDP problem (which has over 2.33 million constraints), thereby creating a new world record. Our implementation also achieved 1.713 PFlops in double precision for large-scale Cholesky factorization using 2,720 CPUs and 4,080 GPUs on the TSUBAME 2.5 supercomputer.

Keywords

Graph analysis Breadth-first search Optimization problem High performance computing Supercomputer Big data 

References

  1. 1.
    Beamer, S., Asanović, K., Patterson, D.A.: Searching for a parent instead of fighting over children: a fast breadth-first search implementation for Graph500. Berkeley, CA: EECS Department, University of California, UCB/EECS-2011-117 (2011)Google Scholar
  2. 2.
    Beamer, S., Asanović, K., Patterson, D.A.: Direction-optimizing breadth-first search. In: Proceedings of the ACM/IEEE International Conference on High Performance Computing, Networking, Storage and Analysis (SC12), IEEE Computer Society (2012)Google Scholar
  3. 3.
    Fujisawa, K., Endo, T., Sato, H., Yamashita, M., Matsuoka, S., Nakata, M.: High-performance general solver for extremely large-scale semidefinite programming problems. In: Proceedings of the 2012 ACM/IEEE Conference on Supercomputing, SC’12 (2012)Google Scholar
  4. 4.
    Fujisawa, K., Endo, T., Sato, H., Yasui, Y., Matsuzawa, N., Waki, H.: Peta-scale general solver for semidefinite programming problems with over two million constraints, SC13 regular, electronic, and educational poster. In: International Conference for High Performance Computing, Networking, Storage and Analysis 2013 (SC2013) (2013)Google Scholar
  5. 5.
    Fujisawa, K., Endo, T., Yasui, Y., Sato, H., Matsuzawa, N., Matsuoka, S., Waki, H.: Peta-scale general solver for semidefinite programming problems with over two million constraints. In: The 28th IEEE International Parallel & Distributed Processing Symposium (IPDPS 2014) (2014)Google Scholar
  6. 6.
    Iwabuchi, K., Sato, H., Mizote, R., Yasui, Y., Fujisawa, K., Matsuoka, S.: Hybrid BFS approach using semi-external memory. In: International Workshop on High Performance Data Intensive Computing (HPDIC2014) in Conjunction with IEEE IPDPS 2014 (2014)Google Scholar
  7. 7.
    Iwabuchi, K., Sato, H., Yasui, Y., Fujisawa, K., Matsuoka, S.: NVM-based hybrid BFS with memory efficient data structure. In: The proceedings of the IEEE BigData2014 (2014)Google Scholar
  8. 8.
    Iwabuchi, K., Sato, H., Yasui, Y., Fujisawa, K.: Performance analysis of hybrid BFS approach using semi-external memory, SC 2013 regular, electronic, and educational poster. In: International Conference for High Performance Computing, Networking, Storage and Analysis (SC2013) (2013)Google Scholar
  9. 9.
    Koch, T., Martin, A., Pfetsch, M.E.: Progress in Academic Computational Integer Programming, Facets of Combinatorial Optimization—Festschrift for Martin Grötschel, pp. 483–506. Springer (2013)Google Scholar
  10. 10.
    Koch, T., Ralphs, T., Shinano, Y.: Could we use a millon cores to solve an integer program? Math. Methods Oper. Res. 76, 67–93 (2012)MATHMathSciNetCrossRefGoogle Scholar
  11. 11.
    Shinano, Y., Achterberg, T., Berthold, T., Heinz, S., Koch, T.: ParaSCIP: a parallel extension of SCIP. In: Competence in High Performance Computing 2010, pp. 135–148. Springer (2012)Google Scholar
  12. 12.
    Suzumura, T., Ueno, K., Sato, H., Fujisawa, K., Matsuoka, S.: A performance characteristics of Graph500 on large-scale distributed environment. In: The proceedings of the 2011 IEEE International Symposium on Workload Characterization (2011)Google Scholar
  13. 13.
    Ueno, K., Suzumura, T.: Highly scalable graph search for the Graph500 benchmark. In: HPDC 2012 (The 21st International ACM Symposium on High-Performance Parallel and Distributed Computing), Delft, Netherlands (2012)Google Scholar
  14. 14.
    Yamashita, M., Fujisawa, K., Fukuda, M., Kobayashi, K., Nakata, K., Nakata, M.: Latest developments in the SDPA family for solving large-scale SDPs. In: Anjos, M.F., Lasserre, J.B. (eds.), Handbook on Semidefinite, Conic and Polynomial Optimization, International Series in Operations Research & Management Science, Chapter 24 (2011)Google Scholar
  15. 15.
    Yamashita, M., Fujisawa, K., Fukuda, M., Nakata, K., Nakata, M.: Parallel solver for semidefinite programming problem having sparse Schur complement matrix. In: ACM Transactions on Mathematical Software, vol. 39, Number 12 (2012)Google Scholar
  16. 16.
    Yasui, Y., Fujisawa, K., Goto, K.: NUMA-optimized parallel breadth-first search on multicore single-node system. In: The Proceedings of the IEEE BigData2013 (2013)Google Scholar
  17. 17.
    Yasui, Y., Fujisawa, K., Goto, K., Kamiyama, N., Takamatsu, M.: NETAL: high-performance implementation of network analysis library considering computer memory hierarchy. J. Oper. Res. Soc. Jpn. 54(4), 259280 (2011)MathSciNetGoogle Scholar
  18. 18.
    Yasui, Y., Fujisawa, K., Sato, Y.: Fast and energy-efficient breadth-first search on a single NUMA system. In: Intentional Supercomputing Conference (ISC 14) (2014)Google Scholar

Copyright information

© Springer Japan 2016

Authors and Affiliations

  • Katsuki Fujisawa
    • 1
  • Toyotaro Suzumura
    • 2
  • Hitoshi Sato
    • 3
  • Koji Ueno
    • 4
  • Yuichiro Yasui
    • 1
  • Keita Iwabuchi
    • 4
  • Toshio Endo
    • 3
  1. 1.Institute of Mathematics for IndustryKyushu UniversityNishi-ku FukuokaJapan
  2. 2.University College DublinDublinIreland
  3. 3.Global Scientific Information and Computing CenterTokyo Institute of TechnologyTokyoJapan
  4. 4.Department of Mathematical and Computing SciencesTokyo Institute of TechnologyTokyoJapan

Personalised recommendations