X10 and APGAS at Petascale

被引:0
|
作者
Tardieu, Olivier [1 ]
Herta, Benjamin [1 ]
Cunningham, David [2 ]
Grove, David [1 ]
Kambadur, Prabhanjan [1 ]
Saraswat, Vijay [1 ]
Shinnar, Avraham [1 ]
Takeuchi, Mikio [3 ]
Vaziri, Mandana [1 ]
机构
[1] IBM Corp, TJ Watson Res Ctr, Armonk, NY 10504 USA
[2] Google Inc, Mountain View, CA USA
[3] IBM Res Tokyo, Tokyo, Japan
关键词
X10; APGAS; scalability; performance;
D O I
10.1145/2692916.2555245
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
X10 is a high-performance, high-productivity programming language aimed at large-scale distributed and shared-memory parallel applications. It is based on the Asynchronous Partitioned Global Address Space (APGAS) programming model, supporting the same fine-grained concurrency mechanisms within and across shared-memory nodes. We demonstrate that X10 delivers solid performance at petascale by running (weak scaling) eight application kernels on an IBM Power 775 supercomputer utilizing up to 55,680 Power7 cores (for 1.7 Pflop/s of theoretical peak performance). We detail our advances in distributed termination detection, distributed load balancing, and use of high-performance interconnects that enable X10 to scale out to tens of thousands of cores. For the four HPC Class 2 Challenge benchmarks, X10 achieves 41% to 87% of the system's potential at scale (as measured by IBM's HPCC Class 1 optimized runs). We also implement K-Means, Smith-Waterman, Betweenness Centrality, and Unbalanced Tree Search (UTS) for geometric trees. Our UTS implementation is the first to scale to petaflop systems.
引用
收藏
页码:53 / 66
页数:14
相关论文
共 50 条
  • [21] 【复古风】 Fujifilm X10
    师毅
    个人电脑, 2012, 18 (02) : 28 - 30
  • [22] Optimizing Remote Data Transfers in X10
    Thangamani, Arun
    Nandivada, V. Krishna
    27TH INTERNATIONAL CONFERENCE ON PARALLEL ARCHITECTURES AND COMPILATION TECHNIQUES (PACT 2018), 2018,
  • [24] MIX10: Compiling MATLAB to X10 for high performance
    1600, Association for Computing Machinery, 2 Penn Plaza, Suite 701, New York, NY 10121-0701, United States (49):
  • [25] MiX10: Compiling MATLAB to X10 for High Performance
    Kumar, Vineet
    Hendren, Laurie
    ACM SIGPLAN NOTICES, 2014, 49 (10) : 617 - 636
  • [26] A practical guide to optimization in X10 expansion microscopy
    Sven Truckenbrodt
    Christoph Sommer
    Silvio O. Rizzoli
    Johann G. Danzl
    Nature Protocols, 2019, 14 : 832 - 863
  • [27] Easy X10 projects for creating a smart home
    Ellis, K
    LIBRARY JOURNAL, 2005, 130 (10) : 172 - 172
  • [28] Home Automation: Insteon (X10 Meets Powerline)
    Lumpkins, William
    IEEE CONSUMER ELECTRONICS MAGAZINE, 2015, 4 (04) : 140 - 144
  • [29] X10程序的差别分析方法
    陈雨亭
    杨威
    赵建军
    计算机学报, 2015, 38 (05) : 1082 - 1092
  • [30] A practical guide to optimization in X10 expansion microscopy
    Truckenbrodt, Sven
    Sommer, Christoph
    Rizzoli, Silvio O.
    Danzl, Johann G.
    NATURE PROTOCOLS, 2019, 14 (03) : 832 - 863