Finding the limits of power-constrained application performance

Peter E. Bailey, Aniruddha Marathe, David K. Lowenthal, Barry Rountree, Martin Schulz

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

34 Scopus citations

Abstract

As we approach exascale systems, power is turning from an optimization goal to a critical operating constraint. With power bounds imposed by both stakeholders and the limitations of existing infrastructure, we need to develop new techniques that work with limited power to extract maximum performance. In this paper, we explore this area and provide an approach to find the theoretical upper bound of computational performance on a per-application basis in hybrid MPI + OpenMP applications. We use a linear programming (LP) formulation to optimize application schedules under various power constraints, where a schedule consists of a DVFS state and number of OpenMP threads for each section of computation between consecutive MPI calls. We also provide a more flexible mixed integer-linear (ILP) formulation and show that the resulting schedules closely match schedules from the LP formulation. Across four applications, we use our LP-derived upper bounds to show that current approaches trail optimal, power-constrained performance by up to 41.1%. This demonstrates the untapped potential of current systems, and our LP formulation provides future optimization approaches with a quantitative optimization target.

Original languageEnglish
Title of host publicationProceedings of SC 2015
Subtitle of host publicationThe International Conference for High Performance Computing, Networking, Storage and Analysis
PublisherIEEE Computer Society
ISBN (Electronic)9781450337236
DOIs
StatePublished - 15 Nov 2015
Externally publishedYes
EventInternational Conference for High Performance Computing, Networking, Storage and Analysis, SC 2015 - Austin, United States
Duration: 15 Nov 201520 Nov 2015

Publication series

NameInternational Conference for High Performance Computing, Networking, Storage and Analysis, SC
Volume15-20-November-2015
ISSN (Print)2167-4329
ISSN (Electronic)2167-4337

Conference

ConferenceInternational Conference for High Performance Computing, Networking, Storage and Analysis, SC 2015
Country/TerritoryUnited States
CityAustin
Period15/11/1520/11/15

Fingerprint

Dive into the research topics of 'Finding the limits of power-constrained application performance'. Together they form a unique fingerprint.

Cite this