Scalable Single Source Shortest Path Algorithms for Massively Parallel Systems

被引:36
作者
Chakaravarthy, Venkatesan T. [1 ]
Checconi, Fabio [2 ]
Murali, Prakash [1 ]
Petrini, Fabrizio [2 ,3 ]
Sabharwal, Yogish [1 ]
机构
[1] IBM Res, New Delhi 110070, India
[2] IBM Thomas J Watson Res Ctr, Yorktown Hts, NY 10598 USA
[3] Intel Parallel Comp Labs, Santa Clara, CA 95054 USA
关键词
Shortest path; parallel algorithm; delta stepping; graph; 500; benchmark; distributed system;
D O I
10.1109/TPDS.2016.2634535
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
We consider the single-source shortest path (SSSP) problem: given an undirected graph with integer edge weights and a source vertex v, find the shortest paths from v to all other vertices. In this paper, we introduce a novel parallel algorithm, derived from the Bellman-Ford and Delta-stepping algorithms. We employ various pruning techniques, such as edge classification and direction-optimization, to dramatically reduce inter-node communication traffic, and we propose load balancing strategies to handle higher-degree vertices. These techniques are particularly effective on power-law graphs, as demonstrated by our extensive performance analysis. In the largest tested configuration, an R-MAT graph with 2(38) vertices and 2(42) edges on 32,768 Blue Gene/Q nodes, we have achieved a processing rate of three Trillion Edges Per Second (TTEPS), a four orders of magnitude improvement over the best published results.
引用
收藏
页码:2031 / 2045
页数:15
相关论文
共 27 条
  • [11] COMPUTATIONAL ANALYSIS OF ALTERNATIVE ALGORITHMS AND LABELING TECHNIQUES FOR FINDING SHORTEST PATH TREES
    DIAL, R
    GLOVER, F
    KARNEY, D
    KLINGMAN, D
    [J]. NETWORKS, 1979, 9 (03) : 215 - 248
  • [12] Edmonds Nick, 2006, 9 DIMACS IMPL CHALL
  • [13] FIBONACCI HEAPS AND THEIR USES IN IMPROVED NETWORK OPTIMIZATION ALGORITHMS
    FREDMAN, ML
    TARJAN, RE
    [J]. JOURNAL OF THE ACM, 1987, 34 (03) : 596 - 615
  • [14] Geisberger R, 2008, LECT NOTES COMPUT SC, V5038, P319, DOI 10.1007/978-3-540-68552-4_24
  • [15] Gonzalez J.E., 2012, 10 USENIX S OP SYST, P17
  • [16] Efficient parallel algorithms for computing all pair shortest paths in directed graphs
    Han, YJ
    Pan, VY
    Reif, JH
    [J]. ALGORITHMICA, 1997, 17 (04) : 399 - 415
  • [17] Harish P, 2007, LECT NOTES COMPUT SC, V4873, P197
  • [18] Querying K-Truss Community in Large and Dynamic Graphs
    Huang, Xin
    Cheng, Hong
    Qin, Lu
    Tian, Wentao
    Yu, Jeffrey Xu
    [J]. SIGMOD'14: PROCEEDINGS OF THE 2014 ACM SIGMOD INTERNATIONAL CONFERENCE ON MANAGEMENT OF DATA, 2014, : 1311 - 1322
  • [19] Kleinberg J, 2006, Algorithm design
  • [20] Leskovec J, 2010, J MACH LEARN RES, V11, P985