A methodology towards automatic performance analysis of parallel applications

被引：5
|
作者：
Calzarossa, M [1 ]
Massari, L [1 ]
Tessera, D [1 ]
机构：
[1] Univ Pavia, Dipartimento Informat & Sistemist, I-27100 Pavia, Italy
来源：
PARALLEL COMPUTING | 2004年 / 30卷 / 02期
关键词：
parallel application; performance analysis; performance metrics; load imbalance;
D O I：
10.1016/j.parco.2003.08.002
中图分类号：
TP301 [理论、方法];
学科分类号：
081202 ;
摘要：
Tuning and debugging the performance of parallel applications is an iterative process consisting of several steps dealing with identification and localization of inefficiencies, repair, and verification of the achieved performance. In this paper, we address the analysis of the performance of parallel applications from a methodological viewpoint with the aim of identifying and localizing inefficiencies. Our methodology is based on performance metrics and criteria that highlight the properties of the applications and the load imbalance and dissimilarities in the behavior of the processors. A few case studies illustrate the application of the methodology. (C) 2003 Elsevier B.V. All rights reserved.
引用
收藏
页码：211 / 223
页数：13
相关论文
共 50 条
  • [1] Towards a communication characterization methodology for parallel applications
    Chodnekar, S
    Srinivasan, V
    Vaidya, AS
    Sivasubramaniam, A
    Das, CR
    [J]. THIRD INTERNATIONAL SYMPOSIUM ON HIGH-PERFORMANCE COMPUTER ARCHITECTURE - PROCEEDINGS, 1997, : 310 - 319
  • [2] Performance assessment of parallel spectral analysis: Towards a practical performance model for parallel medical applications
    Munz, F
    Ludwig, T
    Ziegler, S
    Bartenstein, P
    Schwaiger, M
    Bode, A
    [J]. HIGH-PERFORMANCE COMPUTING AND NETWORKING, PROCEEDINGS, 1999, 1593 : 430 - 439
  • [3] Performance assessment of parallel spectral analysis: towards a practical performance model for parallel medical applications
    Munz, F
    Ludwig, T
    Ziegler, S
    Bartenstein, P
    Schwaiger, M
    Bode, A
    [J]. FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE, 2000, 16 (05): : 553 - 562
  • [4] PERFLOW: A Domain Specific Framework for Automatic Performance Analysis of Parallel Applications
    Jin, Yuyang
    Wang, Haojie
    Zhong, Runxin
    Zhang, Chen
    Zhai, Jidong
    [J]. PPOPP'22: PROCEEDINGS OF THE 27TH ACM SIGPLAN SYMPOSIUM ON PRINCIPLES AND PRACTICE OF PARALLEL PROGRAMMING, 2022, : 177 - 191
  • [5] A METHODOLOGY FOR PERFORMANCE EVALUATION OF PARALLEL APPLICATIONS ON MULTIPROCESSORS
    MENASCE, DA
    BARROSO, LA
    [J]. JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING, 1992, 14 (01) : 1 - 14
  • [6] Automatic analysis of inefficiency patterns in parallel applications
    Wolf, Felix
    Mohr, Bernd
    Dongarra, Jack
    Moore, Shirley
    [J]. CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE, 2007, 19 (11): : 1481 - 1496
  • [7] Automatic performance diagnosis of parallel applications on heterogeneous systems
    Zhan, Kunlin
    Xu, Jungang
    Zhan, Jianfeng
    [J]. International Journal of Digital Content Technology and its Applications, 2012, 6 (02) : 1 - 9
  • [8] GRID-BASED WORKFLOW MANAGEMENT FOR AUTOMATIC PERFORMANCE ANALYSIS OF MASSIVELY PARALLEL APPLICATIONS
    Becker, Daniel
    Riedel, Morris
    Streit, Achim
    Wolft, Felix
    [J]. GRID AND SERVICES EVOLUTION, 2009, : 103 - 117
  • [9] Towards the Performance Investigation of Automatic Melanoma Diagnosis Applications
    Asif, Amna
    Fatima, Iram
    Anjum, Adeel
    Malik, Saif U. R.
    [J]. INTERNATIONAL JOURNAL OF ADVANCED COMPUTER SCIENCE AND APPLICATIONS, 2019, 10 (03) : 390 - 399
  • [10] A performance prediction methodology for data-dependent parallel applications
    Fritzsche, P.
    Roig, C.
    Ripoll, A.
    Luque, E.
    Hernandez, A.
    [J]. 2006 IEEE INTERNATIONAL CONFERENCE ON CLUSTER COMPUTING, VOLS 1 AND 2, 2006, : 381 - +
12345