Automatic scoping of task clauses for the OpenMP tasking model

被引:6
|
作者
Wang, Chun-Kun [1 ]
Chen, Peng-Sheng [2 ]
机构
[1] Natl Chung Cheng Univ, Dept Comp Sci & Informat Engn, Chiayi 621, Taiwan
[2] Natl Chung Cheng Univ, Adv Inst Mfg High Tech Innovat, Dept Comp Sci & Informat Engn, Chiayi 621, Taiwan
来源
JOURNAL OF SUPERCOMPUTING | 2015年 / 71卷 / 03期
关键词
OpenMP; Tasking model; Parallelization; Validation;
D O I
10.1007/s11227-014-1326-3
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
OpenMP provides an easy-to-learn and powerful programming environment for the development of parallel programs. We propose here an algorithm for the automatic correction of the OpenMP tasking model. Assuming a compiler or programmer has identified task regions in the source programs, the proposed algorithm will automatically generate correct task clauses and synchronization. The proposed algorithm is implemented here based on the ROSE compiler infrastructure; 14 benchmark programs are tested, each of which has had all clauses in the task directives removed for the evaluation. The results of this experimental evaluation show that the proposed technique can successfully generate correct clauses for the tested benchmark programs. The proposed technique can simplify the parallelizing of programs using the OpenMP tasking model, making parallel programming more effective and productive.
引用
收藏
页码:808 / 823
页数:16
相关论文
共 50 条
  • [31] Asynchronous OpenMP Tasking with Easy Thread Context Switching and Pool Control
    Fan, Xing
    Sinnen, Oliver
    Giacaman, Nasser
    SCALING OPENMP FOR EXASCALE PERFORMANCE AND PORTABILITY (IWOMP 2017), 2017, 10468 : 217 - 230
  • [32] Performance study of multithreaded MPI and OpenMP tasking in a large scientific code
    Akhmetova, Dana
    Iakymchuk, Roman
    Ekeberg, Orjan
    Laure, Erwin
    2017 IEEE INTERNATIONAL PARALLEL AND DISTRIBUTED PROCESSING SYMPOSIUM WORKSHOPS (IPDPSW), 2017, : 756 - 765
  • [33] How to Reconcile Event-Based Performance Analysis with Tasking in OpenMP
    Lorenz, Daniel
    Mohr, Bernd
    Roessel, Christian
    Schmidl, Dirk
    Wolf, Felix
    BEYOND LOOP LEVEL PARALLELISM IN OPENMP: ACCELERATORS, TASKING AND MORE, PROCEEDINGS, 2010, 6132 : 109 - +
  • [34] Analysis of Explicit vs. Implicit Tasking in OpenMP using Kripke
    Jin, Charles
    Baskaran, Muthu
    PROCEEDINGS OF 2018 IEEE/ACM 4TH INTERNATIONAL WORKSHOP ON EXTREME SCALE PROGRAMMING MODELS AND MIDDLEWARE (ESPM2 2018), 2018, : 62 - 70
  • [35] Automatic Scan Parallelization in OpenMP
    Zegarra, Maicol
    Pereira, Marcio
    Martorell, Xavier
    Araujo, Guido
    2017 INTERNATIONAL SYMPOSIUM ON COMPUTER ARCHITECTURE AND HIGH PERFORMANCE COMPUTING WORKSHOPS (SBAC-PADW), 2017, : 85 - 90
  • [36] Revisiting OpenMP Auto-Scoping Rules
    Dimakopoulos, Vassilios V. V.
    Mourelis, Agelos
    PARALLEL PROCESSING LETTERS, 2023, 33 (01N02)
  • [37] Heterogeneous Task Scheduling for Accelerated OpenMP
    Scogland, Thomas R. W.
    Rountree, Barry
    Feng, Wu-chun
    de Supinski, Bronis R.
    2012 IEEE 26TH INTERNATIONAL PARALLEL AND DISTRIBUTED PROCESSING SYMPOSIUM (IPDPS), 2012, : 144 - 155
  • [38] Variation-tolerant OpenMP Tasking on Tightly-coupled Processor Clusters
    Rahimi, Abbas
    Marongiu, Andrea
    Burgio, Paolo
    Gupta, Rajesh K.
    Benini, Luca
    DESIGN, AUTOMATION & TEST IN EUROPE, 2013, : 541 - 546
  • [39] Evaluation of OpenMP task scheduling strategies
    Duran, Alejandro
    Corbalan, Julita
    Ayguade, Eduard
    OPENMP IN A NEW ERA OF PARALLELISM, PROCEEDINGS, 2008, 5004 : 100 - 110
  • [40] OpenMP Task Scheduling Analysis via OpenMP Runtime API and Tool Visualization
    Qawasmeh, Ahmad
    Malik, Abid
    Chapman, Barbara
    PROCEEDINGS OF 2014 IEEE INTERNATIONAL PARALLEL & DISTRIBUTED PROCESSING SYMPOSIUM WORKSHOPS (IPDPSW), 2014, : 1050 - 1059