MetaCartSign in to MyCiteSeer

Include Citations | Advanced Search | Help

Include Citations | Advanced Search | Help

  Cause-Effect Analysis of Parallel Program Performance

Download:
Download as a PDF | Download as a PS
by Wagner Meira, Thomas J. Leblanc
ftp://ftp.cs.unc.edu/pub/users/sc/ppopp97/17.ps.Z
Add To MetaCart

Abstract:

This paper describes a general framework and several specific techniques for cause-effect analysis: an automated inference process that presents explanations for dynamic phenomena of parallel program executions in terms of underlying causes and the related source code. We illustrate the framework by describing the implementation of three analysis techniques: waiting time analysis identifies the cause of synchronization overhead as the differences in execution paths taken by synchronizing processors; protocol analysis identifies the sharing patterns to pages that produce invalidations in a DSM protocol; and transaction analysis identifies conflicts between transactions in a parallel file system that cause aborts. We present examples of how each technique can be used to understand observed performance effects, and how insights derived from these techniques suggest program modifications that lead to improved performance. 1

Citations

1746 Time, clocks, and the ordering of events in a distributed system – Lamport - 1978
724 The SPLASH-2 programs: Characterization and methodological considerations – Woo, Ohara, et al. - 1995
705 SPLASH: Stanford Parallel Applications for Shared Memory – Singh, Weber, et al. - 1992
372 TreadMarks: Shared Memory Computing on Networks of Workstations – Amza, Cox, et al. - 1996
87 MemSpy: Analyzing memory system bottlenecks in programs – MARTONOSI, GUPTA, et al. - 1992
66 Mtool: An integrated system for performance debugging shared memory multiprocessor applications – Goldberg, Hennessy - 1993
54 An Integrated Compilation and Performance Analysis Environment for Data Parallel Programs," presented at Supercomputing '95 – Adve, Mellor-Crummey, et al. - 1995
51 Parallel Data Mining for Association Rules on Shared-Memory Systems – Parthasarathy, Zaki, et al.
12 StormWatch: A Tool for Visualizing Memory System Protocols – Chilimbi, Ball, et al. - 1995
12 Waiting time analysis and performance visualization in Carnival – Jr, LeBlanc, et al. - 1996
10 A performance debugger for eliminating excess synchronization in shared-memory parallel programs – Rajamony, Cox - 1996
1 Understanding the performance of DSM applications – Jr, LeBlanc, et al. - 1997
1 PIOUS for PVM: User's guide and reference manual -- version 1 – Moyer, Sunderan - 1995
1 Shared-memory performance profiling. Available at ftp://grilled.cs.wisc.edu/technical papers/ppopp97.ps.Z – Xu, Larus, et al. - 1996