Representation of coherency classes for parallel systems

Walter Hussak, John A. Keane

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Abstract

    Some parallel applications do not require a precise imitation of the behaviour of the physically shared memory programming model. Consequently, certain parallel machine architectures have elected to emphasize different required coherency properties because of possible efficiency gains. This has led to various definitions of models of store coherency. These definitions have not been amenable to detailed analysis and, consequently, inconsistencies have resulted. In this paper a unified framework is proposed in which different models of store coherency are developed systematically by progressively relaxing the constraints that they have to satisfy. A demonstration is given of how formal reasoning can be carried out to compare different models. Some real-life systems are considered and a definition of a version of weak coherency is found to be incomplete.
    Original languageEnglish
    Title of host publicationProceedings of the 5th IEEE Symposium on Parallel and Distributed Processing|Proc 5 IEEE Symp Parallel Distrib Proc
    Editors Anon
    Place of PublicationLos Alamitos, CA, United States
    PublisherIEEE
    Pages391-398
    Number of pages7
    ISBN (Print)081864222X
    Publication statusPublished - 1993
    EventProceedings of the 5th IEEE Symposium on Parallel and Distributed Processing - Dallas, TX, USA
    Duration: 1 Jul 1993 → …
    http://dblp.uni-trier.de/db/conf/spdp/spdp1993.html#HussakK93http://dblp.uni-trier.de/rec/bibtex/conf/spdp/HussakK93.xmlhttp://dblp.uni-trier.de/rec/bibtex/conf/spdp/HussakK93

    Conference

    ConferenceProceedings of the 5th IEEE Symposium on Parallel and Distributed Processing
    CityDallas, TX, USA
    Period1/07/93 → …
    Internet address

    Fingerprint

    Dive into the research topics of 'Representation of coherency classes for parallel systems'. Together they form a unique fingerprint.

    Cite this