MP-STREAM: A Memory Performance Benchmark for Design Space Exploration on Heterogeneous HPC Devices

Nabi, S. W. and Vanderbauwhede, W. (2018) MP-STREAM: A Memory Performance Benchmark for Design Space Exploration on Heterogeneous HPC Devices. In: 32nd IEEE International Parallel and Distributed Processing Symposium, Reconfigurable Architectures Workshop (RAW 2018), Vancouver, BC, Canada, 21-25 May 2018, ISBN 9781538655559 (doi:10.1109/IPDPSW.2018.00036)

[img]
Preview
Text
158697.pdf - Accepted Version

378kB

Abstract

Sustained memory throughput is a key determinant of performance in HPC devices. Having an accurate estimate of this parameter is essential for manual or automated design space exploration for any HPC device. While there are benchmarks for measuring the sustained memory bandwidth for CPUs and GPUs, such a benchmark for FPGAs has been missing. We present MP-STREAM, an OpenCL-based synthetic micro-benchmark for measuring sustained memory bandwidth, optimized for FPGAs, but which can be used on multiple platforms. Our main contribution is the introduction of various generic as well as device-specific parameters that can be tuned to measure their effect on memory bandwidth. We present results of running our benchmark on a CPU, a GPU and two FPGA targets, and discuss our observations. The experiments underline the utility of our benchmark for optimizing HPC applications for FPGAs, and provide valuable optimization hints for FPGA programmers.

Item Type:Conference Proceedings
Status:Published
Refereed:Yes
Glasgow Author(s) Enlighten ID:Vanderbauwhede, Dr Wim and Nabi, Dr Syed Waqar
Authors: Nabi, S. W., and Vanderbauwhede, W.
College/School:College of Science and Engineering > School of Computing Science
ISBN:9781538655559
Copyright Holders:Copyright © 2018 IEEE
Publisher Policy:Reproduced in accordance with the copyright policy of the publisher
Related URLs:

University Staff: Request a correction | Enlighten Editors: Update this record

Project CodeAward NoProject NamePrincipal InvestigatorFunder's NameFunder RefLead Dept
614451Exploiting Parallelism through Type Transformations for Hybrid Manycore Systems.Wim VanderbauwhedeEngineering and Physical Sciences Research Council (EPSRC)EP/L00058X/1COM - COMPUTING SCIENCE