Davis, J. A., Mudalige, Gihan R., Hammond, Simon D., Herdman, J. A., Miller, I. and Jarvis, Stephen A. (2011) Predictive analysis of a hydrodynamics application on large-scale CMP clusters. Computer Science - Research and Development, Volume 26 (Number 3-4). pp. 175-185. doi:10.1007/s00450-011-0164-2

Preview

PDF WRAP_Herdman_Hydra.pdf - Accepted Version - Requires a PDF viewer. Download (577Kb) | Preview

Request Changes to record.

Abstract

We present the development of a predictive performance model for the high-performance computing code Hydra, a hydrodynamics benchmark developed and maintained by the United Kingdom Atomic Weapons Establishment (AWE). The developed model elucidates the parallel computation of Hydra, with which it is possible to predict its runtime and scaling performance on varying large-scale chip multiprocessor (CMP) clusters. A key feature of the model is its granularity; with the model we are able to separate the contributing costs, including computation, point-to-point communications, collectives, message buffering and message synchronisation. The predictions are validated on two contrasting large-scale HPC systems, an AMD Opteron/ InfiniBand cluster and an IBM BlueGene/P, both of which are located at the Lawrence Livermore National Laboratory (LLNL) in the US. We validate the model on up to 2,048 cores, where it achieves a > 85% accuracy in weak-scaling studies. We also demonstrate use of the model in exposing the increasing costs of collectives for this application, and also the influence of node density on network accesses, therefore highlighting the impact of machine choice when running this hydrodynamics application at scale.

Item Type:	Journal Article
Subjects:	Q Science > QA Mathematics > QA76 Electronic computers. Computer science. Computer software
Divisions:	Faculty of Science > Computer Science
Library of Congress Subject Headings (LCSH):	High performance computing, Hydrodynamics -- Computer programs
Journal or Publication Title:	Computer Science - Research and Development
Publisher:	Springer
Place of Publication:	Hamburg, Germany
ISSN:	1865-2034
Official Date:	June 2011
Dates:	Date Event June 2011 UNSPECIFIED
Date of first compliant deposit:	21 December 2015
Volume:	Volume 26
Number:	Number 3-4
Number of Pages:	11
Page Range:	pp. 175-185
DOI:	10.1007/s00450-011-0164-2
Status:	Peer Reviewed
Publication Status:	Published
Access rights to Published version:	Restricted or Subscription Access
Funder:	Atomic Weapons Establishment (Great Britain) (AWE), University of Warwick. Centre for Scientific Computing, Science Research Investment Fund (SRIF)
Grant number:	CDK0660 (AWE), CDK0724 (AWE)
Conference Paper Type:	Paper
Title of Event:	Proceedings of the International Supercomputing Conference (ISC'11)
Type of Event:	Conference
Location of Event:	Hamburg, Germany
Date(s) of Event:	19-23rd June 2011

Request changes or add full text files to a record

Repository staff actions (login required)

View Item

Downloads