Realistic evaluation of interconnection network performance at high loads

F. J. Ridruejo, J. Navaridas, J. Miguel-Alonso, Cruz Izu

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

    Abstract

    Any simulation-based evaluation of an interconnection network proposal requires a good characterization of the workload. Synthetic traffic patterns based on independent traffic sources are commonly used to measure performance in terms of average latency and peak throughput. As they do not capture the level of self-throttling that occurs in most parallel applications, they can produce inaccurate throughput estimates at high loads. Thus, workloads that resemble the varying levels of synchronization of actual applications are needed to study the performance of interconnection networks. One approach is to use simple, burst-synchronized synthetic workloads that emulate the self-throttling of many parallel applications. To validate this approach, we compare the gains achieved by a restrictive injection mechanism under this workload with those obtained using traces from the NAS Parallel Benchmarks. This study confirms that the burst-synchronized traffic model provides reasonable performance estimates, which could be improved by taking into account dependency chains between messages. © 2007 IEEE.
    Original languageEnglish
    Title of host publicationParallel and Distributed Computing, Applications and Technologies, PDCAT Proceedings|Parallel Distrib. Comput. Appl. Technol. PDCAT Proc.
    Pages97-104
    Number of pages7
    DOIs
    Publication statusPublished - 2007
    Event18th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2007 - Adelaide, SA
    Duration: 1 Jul 2007 → …

    Conference

    Conference18th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2007
    CityAdelaide, SA
    Period1/07/07 → …

    Fingerprint

    Dive into the research topics of 'Realistic evaluation of interconnection network performance at high loads'. Together they form a unique fingerprint.

    Cite this