High Performance and Enhanced Scalability for Parallel Applications using MPI-3's non-blocking Collectives

Pericherla, Surendra Varma and Vadhiyar, Sathish (2017) High Performance and Enhanced Scalability for Parallel Applications using MPI-3's non-blocking Collectives. In: International Conference on Computational Science (ICCS), JUN 12-14, 2017, Zurich, SWITZERLAND, pp. 2403-2407.

PDF Int_Con_Com_Sci_108_2403_2017.pdf - Published Version Restricted to Registered users only Download (483kB) | Request a copy

Abstract

Collective communications occupy 20-90% of total execution times in many MPI applications. In this paper, we propose strategies for automatically identifying the most time-consuming collective operations that also act as scalability bottlenecks. We then explore the use of MPI-3's non-blocking collectives for these communications. We also rearrange the codes to adequately overlap the independent computations with the non-blocking collective communications. Applying these strategies for different graph and machine learning applications, we obtained up to 33% performance improvements for large-scale runs on a Cray supercomputer. (C) 2017 The Authors. Published by Elsevier B.V.

Item Type:	Conference Proceedings
Series.:	Procedia Computer Science
Additional Information:	Copy right for this article belongs to the ELSEVIER SCIENCE BV, SARA BURGERHARTSTRAAT 25, PO BOX 211, 1000 AE AMSTERDAM, NETHERLANDS
Department/Centre:	Division of Interdisciplinary Sciences > Supercomputer Education & Research Centre
Date Deposited:	12 Aug 2017 06:56
Last Modified:	12 Aug 2017 06:56
URI:	http://eprints.iisc.ac.in/id/eprint/57631

Actions (login required)

View Item