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 conferenceOtherOther

    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

    Dinh, M. N., Sadayappan, P. (Ed.), Abramson, D. A., Jin, C., Gontarek, A., Moench, R., & DeRose, L. (2012). Scalable parallel debugging with statistical assertions. 311 - 312. Principles and Practice of Parallel Programming 2012, New York NY USA, . https://doi.org/10.1145/2370036.2145870
    Dinh, Minh Ngoc ; Sadayappan, P (Editor) ; Abramson, David Andrew ; Jin, Chao ; Gontarek, Andrew ; Moench, Robert ; DeRose, Luiz. / Scalable parallel debugging with statistical assertions. Principles and Practice of Parallel Programming 2012, New York NY USA, .2 p.
    @conference{42bf2cceba8d4b4a90395d80d660459d,
    title = "Scalable parallel debugging with statistical assertions",
    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.",
    keywords = "Assertion, Parallel debugging, Statistic",
    author = "Dinh, {Minh Ngoc} and P Sadayappan and Abramson, {David Andrew} and Chao Jin and Andrew Gontarek and Robert Moench and Luiz DeRose",
    year = "2012",
    doi = "10.1145/2370036.2145870",
    language = "English",
    pages = "311 -- 312",
    note = "Principles and Practice of Parallel Programming 2012 ; Conference date: 25-02-2012 Through 29-02-2012",

    }

    Dinh, MN, Sadayappan, P (ed.), Abramson, DA, Jin, C, Gontarek, A, Moench, R & DeRose, L 2012, 'Scalable parallel debugging with statistical assertions' Principles and Practice of Parallel Programming 2012, New York NY USA, 25/02/12 - 29/02/12, pp. 311 - 312. https://doi.org/10.1145/2370036.2145870

    Scalable parallel debugging with statistical assertions. / Dinh, Minh Ngoc; Sadayappan, P (Editor); Abramson, David Andrew; Jin, Chao; Gontarek, Andrew; Moench, Robert; DeRose, Luiz.

    2012. 311 - 312 Principles and Practice of Parallel Programming 2012, New York NY USA, .

    Research output: Contribution to conferenceOtherOther

    TY - CONF

    T1 - Scalable parallel debugging with statistical assertions

    AU - Dinh, Minh Ngoc

    AU - Abramson, David Andrew

    AU - Jin, Chao

    AU - Gontarek, Andrew

    AU - Moench, Robert

    AU - DeRose, Luiz

    A2 - Sadayappan, P

    PY - 2012

    Y1 - 2012

    N2 - 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.

    AB - 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.

    KW - Assertion

    KW - Parallel debugging

    KW - Statistic

    U2 - 10.1145/2370036.2145870

    DO - 10.1145/2370036.2145870

    M3 - Other

    SP - 311

    EP - 312

    ER -

    Dinh MN, Sadayappan P, (ed.), Abramson DA, Jin C, Gontarek A, Moench R et al. Scalable parallel debugging with statistical assertions. 2012. Principles and Practice of Parallel Programming 2012, New York NY USA, . https://doi.org/10.1145/2370036.2145870