Efficient pattern search in large traces through successive refinement

Felix Wolf, Bernd Mohr, Jack Dongarra, Shirley Moore

    Research output: Chapter in Book/Conference proceedingConference contribution

    Abstract

    Event tracing is a well-accepted technique for post-mortem performance analysis of parallel applications. The EXPERT tool supports the analysis of large traces by automatically searching them for execution patterns that indicate inefficient behavior. However, the current search algorithm works with independent pattern specifications and ignores the specialization hierarchy existing between them, resulting in a long analysis time caused by repeated matching attempts as well as in replicated code. This article describes an optimized design taking advantage of specialization relationships and leading to a significant runtime improvement as well as to more compact pattern specifications. © Springer-Verlag 2004.
    Original languageEnglish
    Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|Lect. Notes Comput. Sci.
    PublisherSpringer Nature
    Pages47-54
    Number of pages7
    Volume3149
    Publication statusPublished - 2004
    EventEuro-Par 2004 Parallel Processing, 10th International Euro-Par Conference, Pisa, Italy, August 31-September 3, 2004, Proceedings -
    Duration: 1 Jan 1824 → …
    http://dblp.uni-trier.de/db/conf/europar/europar2004.html#WolfMDM04http://dblp.uni-trier.de/rec/bibtex/conf/europar/WolfMDM04.xmlhttp://dblp.uni-trier.de/rec/bibtex/conf/europar/WolfMDM04

    Publication series

    NameLecture Notes in Computer Science

    Conference

    ConferenceEuro-Par 2004 Parallel Processing, 10th International Euro-Par Conference, Pisa, Italy, August 31-September 3, 2004, Proceedings
    Period1/01/24 → …
    Internet address

    Fingerprint

    Dive into the research topics of 'Efficient pattern search in large traces through successive refinement'. Together they form a unique fingerprint.

    Cite this