论文信息 - A systematic multi-step methodology for performance analysis of communication traces of distributed applications based on hierarchical clustering

A systematic multi-step methodology for performance analysis of communication traces of distributed applications based on hierarchical clustering

Often parallel scientific applications are instrumented and traces are collected and analyzed to identify processes with performance problems or operations that cause delays in program execution. The execution of instrumented codes may generate large amounts of performance data, and the collection, storage, and analysis of such traces are time and space demanding. To address this problem, this paper presents an efficient, systematic, multi-step methodology, based on hierarchical clustering, for analysis of communication traces of parallel scientific applications. The methodology is used to discover potential communication performance problems of three applications: TRACE, REMO, and SWEEP3D

[1] Bernd Mohr,et al. TAU: A Portable Parallel Program Analysis Environment for pC++ , 1994, CONPAR.

[2] Bernd Mohr,et al. Efficient Pattern Search in Large Traces Through Successive Refinement , 2004, Euro-Par.

[3] Ecmwf Newsletter,et al. EUROPEAN CENTRE FOR MEDIUM-RANGE WEATHER FORECASTS , 2004 .

[4] Jeffrey S. Vetter,et al. Scalable Analysis Techniques for Microprocessor Performance Counter Metrics , 2002, ACM/IEEE SC 2002 Conference (SC'02).

[5] Felix Wolf,et al. Automatic performance analysis on parallel computers with SMP nodes , 2003 .

[6] Bernd Mohr,et al. KOJAK - A Tool Set for Automatic Performance Analysis of Parallel Programs , 2003, Euro-Par.

[7] B. Miller,et al. The Paradyn Parallel Performance Measurement Tools , 1995 .

[8] Patricia J. Teller,et al. Accuracy of Performance Monitoring Hardware , 2002 .

[9] Barton P. Miller,et al. The Paradyn Parallel Performance Measurement Tool , 1995, Computer.

[10] Jeffrey S. Vetter,et al. Statistical scalability analysis of communication operations in distributed applications , 2001, PPoPP '01.

[11] Jeffrey S. Vetter,et al. Communication characteristics of large-scale scientific applications for contemporary cluster architectures , 2002, Proceedings 16th International Parallel and Distributed Processing Symposium.

[12] Felix Wolf. EARL - API Documentation , 2004 .