The Library
Predictive evaluation of partitioning algorithms through runtime modelling
Tools
Bunt, Richard A., Wright, Steven A., Jarvis, Stephen A., Street, Matthew and Ho, Yoon K. (2016) Predictive evaluation of partitioning algorithms through runtime modelling. In: High Performance Computing, Data, and Analytics (HiPC'16), Hyderabad, India, 19-22 Dec 2016. Published in: 2016 IEEE 23rd International Conference on High Performance Computing (HiPC) pp. 351-361. ISBN 9781509054121.
PDF
WRAP_081016-model_extension_paper.pdf - Accepted Version - Requires a PDF viewer. Download (585Kb) |
Official URL: http://doi.org/10.1109/HiPC.2016.048
Abstract
Performance modelling unstructured mesh codes is a challenging process, due to the difficulty of capturing their memory access patterns, and their communication patterns at varying scale. In this paper we first develop extensions to an existing runtime performance model, aimed at overcoming the former, which we validate on up to 1,024 cores of a Haswellbased cluster, using both a geometric partitioning algorithm and ParMETIS to partition the input deck, with a maximum absolute runtime error of 12.63% and 11.55% respectively. To overcome the latter, we develop an application representative of the mesh partitioning process internal to an unstructured mesh code. This application is able to generate partitioning data that is usable with the performance model to produce predicted application runtimes within 7.31% of those produced using empirically collected data. We then demonstrate the use of the performance model by undertaking a predictive comparison among several partitioning algorithms on up to 30,000 cores. Additionally, we correctly predict the ineffectiveness of the geometric partitioning algorithm at 512 and 1024 cores.
Item Type: | Conference Item (Paper) | ||||||
---|---|---|---|---|---|---|---|
Subjects: | Q Science > QA Mathematics > QA76 Electronic computers. Computer science. Computer software | ||||||
Divisions: | Faculty of Science, Engineering and Medicine > Science > Computer Science | ||||||
Library of Congress Subject Headings (LCSH): | High performance computing, Fluid dynamics | ||||||
Journal or Publication Title: | 2016 IEEE 23rd International Conference on High Performance Computing (HiPC) | ||||||
Publisher: | IEEE | ||||||
ISBN: | 9781509054121 | ||||||
Official Date: | 2016 | ||||||
Dates: |
|
||||||
Page Range: | pp. 351-361 | ||||||
Status: | Peer Reviewed | ||||||
Publication Status: | Published | ||||||
Access rights to Published version: | Restricted or Subscription Access | ||||||
Date of first compliant deposit: | 10 October 2016 | ||||||
Date of first compliant Open Access: | 10 October 2016 | ||||||
Funder: | Rolls-Royce Group plc, Bull (Firm) | ||||||
Conference Paper Type: | Paper | ||||||
Title of Event: | High Performance Computing, Data, and Analytics (HiPC'16) | ||||||
Type of Event: | Conference | ||||||
Location of Event: | Hyderabad, India | ||||||
Date(s) of Event: | 19-22 Dec 2016 |
Request changes or add full text files to a record
Repository staff actions (login required)
View Item |
Downloads
Downloads per month over past year