A hybrid approach to parallel pattern discovery in C++

Christopher Mark Brown, Vladimir Janjic, Adam David Barwell, John Donald Thomson, Roberto Castañeda Lozano, Murray Cole, Bjoern Franke, Jose Daniel Garcia-Sanchez, David Del Rio Astorga, K. MacKenzie

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)
14 Downloads (Pure)

Abstract

Parallel pattern libraries offer a strong combination of abstraction and performance. However, discovering places in sequential code where parallel patterns should be introduced is still highly non-trivial, often requiring expert manual analysis and profiling. We present a hybrid discovery technique to detect instances of parallel patterns in sequential code. This employs both static and dynamic trace-based analysis, together with hotspot detection. We evaluate our pattern discovery mechanism on a number of representative benchmarks. We evaluate the performance of the resulting parallelised benchmarks on a 24-core parallel machine.
Original languageEnglish
Title of host publication2020 28th Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP)
PublisherIEEE Computer Society
Number of pages5
ISBN (Electronic)9781728165820
ISBN (Print)9781728165837
DOIs
Publication statusPublished - 14 May 2020
Event28th Euromicro International Conference on Parallel, Distributed and Network-based Processing - Sweeden, Västerås, Sweden
Duration: 11 Mar 202013 Mar 2020
Conference number: 28
http://www.pdp2020.com/

Publication series

NameProceedings - Euromicro Workshop on Parallel and Distributed Processing
PublisherIEEE
ISSN (Print)1066-6192
ISSN (Electronic)2377-5750

Conference

Conference28th Euromicro International Conference on Parallel, Distributed and Network-based Processing
Abbreviated titlePDP 2020
Country/TerritorySweden
CityVästerås
Period11/03/2013/03/20
Internet address

Fingerprint

Dive into the research topics of 'A hybrid approach to parallel pattern discovery in C++'. Together they form a unique fingerprint.

Cite this