Scalable parallel debugging with statistical assertions

Minh Ngoc Dinh, P Sadayappan (Editor), David Andrew Abramson, Chao Jin, Andrew Gontarek, Robert Moench, Luiz DeRose

    Research output: Contribution to conferenceOther

    Abstract

    Traditional debuggers are of limited value for modern scientific codes that manipulate large complex data structures. This paper discusses a novel debug-time assertion, called a "Statistical Assertion", that allows a user to reason about large data structures, and the primitives are parallelised to provide an efficient solution. We present the design and implementation of statistical assertions, and illustrate the debugging technique with a molecular dynamics simulation. We evaluate the performance of the tool on a 12,000 cores Cray XE6.

    Original languageEnglish
    Pages311 - 312
    Number of pages2
    DOIs
    Publication statusPublished - 2012
    EventPrinciples and Practice of Parallel Programming 2012 - New Orleans LA USA, New York NY USA
    Duration: 25 Feb 201229 Feb 2012

    Conference

    ConferencePrinciples and Practice of Parallel Programming 2012
    CityNew York NY USA
    Period25/02/1229/02/12

    Keywords

    • Assertion
    • Parallel debugging
    • Statistic

    Cite this