Review of Performance Analysis Tools for MPI Parallel Programs

In order to produce MPI applications that perform well on today’s parallel architectures, programmers need effective tools for collecting and analyzing performance data. Because programmers typically work on more than one platform, cross-platform tools are highly desirable. A variety of such tools, both commercial and research, are becoming available. Because the tools use different approaches and provide different features, the user may be confused as to what tool will best suit his needs. This paper reviews the available cross-platform MPI performance analysis tools and evaluates them according to a set of criteria that includes robustness, usability, scalability, portability, and versatility. Issues pertaining to hybrid and heterogeneous, distributed programming environments and to MPI I/O performance analysis are also discussed.

• Type: Conference
• Year: 2001
• Where: PVM
• Authors: Shirley Moore, David Cronk, Kevin S. London, Jack Dongarra
• Keywords: Cross-platform Mpi Performance, Cross-platform Tools, Distributed And Parallel Computing, PVM 2001, Tools