Performance models for cluster-enabled OpenMP implementations

Jie Cai*, Alistair P. Rendell, Peter E. Strazdins, H'sien Jin Wong

*Corresponding author for this work

    Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

    4 Citations (Scopus)

    Abstract

    A key issue for Cluster-enabled OpenMP implementations based on software Distributed Shared Memory (sDSM) systems, is maintaining the consistency of the shared memory space. This forms the major source of overhead for these systems, and is driven by the detection and servicing of page faults. This paper investigates how application performance can be modelled based on the number of page faults. Two simple models are proposed, one based on the number of page faults along the critical path of the computation, and one based on the aggregated numbers of page faults. Two different sDSM systems are considered. The models are evaluated using the OpenMP NAS Parallel Benchmarks on an 8-node AMD-based Gigabit Ethernet cluster. Both models gave estimates accurate to within 10% in most cases, with the critical path model showing slightly better accuracy; accuracy is lost if the underlying page faults cannot be overlapped, or if the application makes extensive use of the OpenMP flush directive.

    Original languageEnglish
    Title of host publication13th IEEE Asia-Pacific Computer Systems Architecture Conference, ACSAC 2008
    DOIs
    Publication statusPublished - 2008
    Event13th IEEE Asia-Pacific Computer Systems Architecture Conference, ACSAC 2008 - Hsinchu, Taiwan
    Duration: 4 Aug 20086 Aug 2008

    Publication series

    Name13th IEEE Asia-Pacific Computer Systems Architecture Conference, ACSAC 2008

    Conference

    Conference13th IEEE Asia-Pacific Computer Systems Architecture Conference, ACSAC 2008
    Country/TerritoryTaiwan
    CityHsinchu
    Period4/08/086/08/08

    Fingerprint

    Dive into the research topics of 'Performance models for cluster-enabled OpenMP implementations'. Together they form a unique fingerprint.

    Cite this