A methodology for assessing computation/communication overlap of MPI nonblocking collectives. (5th August 2022)

• Record Type: Journal Article
• Authors: Denis, Alexandre; Jaeger, Julien; Jeannot, Emmanuel; Reynier, Florian
• Is Part Of: Concurrency and computation. Volume 34:Number 22(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-08-05
• Subjects: collective communication -- high‐performance computing -- MPI; Parallel processing (Electronic computers) -- Periodicals; Parallel computers -- Periodicals; 004.35
• DOI: 10.1002/cpe.7168
• Languages: English
• ISSNs: 1532-0626

Abstract

Summary: By allowing computation/communication overlap, MPI nonblocking collectives (NBC) are supposed to improve application scalability and performance. However, it is known that to actually get overlap, the MPI library has to implement progression mechanisms in software or rely on the network hardware. These mechanisms may be present or not, adequate or perfectible, they may have an impact on communication performance or may interfere with computation by stealing CPU cycles. From a user point of view, assessing and understanding the behavior of an MPI library concerning computation/communication overlap is difficult. In this article, we propose a methodology to assess the computation/communication overlap of NBC. We propose new metrics to measure how much communication and computation do overlap, and to evaluate how they interfere with each other. We integrate these metrics into a complete methodology. We compare our methodology with state of the art metrics and benchmarks, and show that ours provides more meaningful informations. We perform experiments on a large panel of MPI implementations and network hardware and show when and why overlap is efficient, nonexistent or even degrades performance.