DocumentCode
2034190
Title
Optimizing Parallel Programs Using Affinity Regions
Author
Appelbe, Bill ; Lakshmanan, Batakrishnan
Author_Institution
Georgia Institue of Technology, Atlanta, GA
Volume
2
fYear
1993
fDate
16-20 Aug. 1993
Firstpage
246
Lastpage
249
Abstract
Afinity regions ensure that a shared processor scheduling, mapping iterations to processors, is used in consecutive parallel loop nests. Using affinity regions can significantly improve locality of reference and hence the performance. This paper discusses the use and effectiveness of affinity regions.
Keywords
Algorithm design and analysis; Concurrent computing; Educational institutions; Electronic mail; Load management; Merging; Parallel processing; Processor scheduling; Programming profession; Yarn;
fLanguage
English
Publisher
ieee
Conference_Titel
Parallel Processing, 1993. ICPP 1993. International Conference on
Conference_Location
Syracuse, NY, USA
ISSN
0190-3918
Print_ISBN
0-8493-8983-6
Type
conf
DOI
10.1109/ICPP.1993.131
Filename
4134217
Link To Document