In EDS ansehen

Holistic Hardware Counter Performance Analysis of Parallel Programs

Gespeichert in:
Bibliographische Detailangaben
Titel: Holistic Hardware Counter Performance Analysis of Parallel Programs
Autoren: Wylie, Brian J. N., Mohr, Bernd, Wolf, Felix
Weitere Verfasser: Brian J. N. Wylie and Bernd Mohr and Felix Wolf
Verlagsinformationen: Schloss Dagstuhl – Leibniz-Zentrum für Informatik
Publikationsjahr: 2006
Bestand: DROPS - Dagstuhl Research Online Publication Server (Schloss Dagstuhl - Leibniz Center for Informatics )
Schlagwörter: Parallel program execution tracing, hardware performance counters, automated structured analysis, performance comparison
Beschreibung: The KOJAK toolkit has been augmented with refined hardware performance counter support, including more convenient measurement specification, additional metric derivations and hierarchical structuring, and an extended algebra for integrating multiple experiments. Comprehensive automated analysis of a hybrid OpenMP/MPI parallel program, the ASC Purple sPPM benchmark, is demonstrated with performance experiments on equisized POWER4-II-based IBM Regatta p690+ cluster, Opteron-based Cray XD1 cluster and UltraSPARC-IV-based Sun Fire E25000 systems. Automatically assessed communication and synchronisation performance properties, combined with a rich set of measured and derived counter metrics, provide a holistic analysis context and facilitate multi-platform comparison.
Publikationsart: article in journal/newspaper
conference object
Dateibeschreibung: application/pdf
Sprache: English
Relation: Is Part Of Dagstuhl Seminar Proceedings, Volume 5501, Automatic Performance Analysis (2006); https://drops.dagstuhl.de/entities/document/10.4230/DagSemProc.05501.4
DOI: 10.4230/DagSemProc.05501.4
Verfügbarkeit: https://doi.org/10.4230/DagSemProc.05501.4
https://nbn-resolving.org/urn:nbn:de:0030-drops-5071
https://drops.dagstuhl.de/entities/document/10.4230/DagSemProc.05501.4
Rights: https://creativecommons.org/licenses/by/4.0/legalcode
Dokumentencode: edsbas.1F21D80E
Datenbank: BASE
Beschreibung
Abstract:The KOJAK toolkit has been augmented with refined hardware performance counter support, including more convenient measurement specification, additional metric derivations and hierarchical structuring, and an extended algebra for integrating multiple experiments. Comprehensive automated analysis of a hybrid OpenMP/MPI parallel program, the ASC Purple sPPM benchmark, is demonstrated with performance experiments on equisized POWER4-II-based IBM Regatta p690+ cluster, Opteron-based Cray XD1 cluster and UltraSPARC-IV-based Sun Fire E25000 systems. Automatically assessed communication and synchronisation performance properties, combined with a rich set of measured and derived counter metrics, provide a holistic analysis context and facilitate multi-platform comparison.
DOI:10.4230/DagSemProc.05501.4