DocumentCode
1371631
Title
Eliminating memory for fragmentation within partitionable SIMD/SPMD machines
Author
Nichols, Mark A. ; Siegel, Howard Jay ; Dietz, Henry G. ; Quong, Russell W. ; Nation, Wayne G.
Author_Institution
NCR, San Diego, CA, USA
Volume
2
Issue
3
fYear
1991
fDate
7/1/1991 12:00:00 AM
Firstpage
290
Lastpage
303
Abstract
Efficient data layout is an important aspect of the compilation process. A model for the creation of perfect memory maps for large-scale parallel machines capable of user-controlled partitionable single-instruction-multiple data/single-program-multiple data (SIMD/SPMD) operation is developed. The term perfect implies that no memory fragmentation occurs and ensures that the memory map size is kept to a minimum. A major constraint on solving this problem is based on the single program nature of both the SIMD and SPMD modes of parallelism. It is assumed that all processors within the same submachine used identical addresses to access corresponding data items in each of their local memories. Necessary and sufficient conditions are derived for being able to create perfect memory maps, and results are applied to several partitionable interconnection networks
Keywords
multiprocessor interconnection networks; parallel machines; parallel programming; program compilers; storage management; compilation process; large-scale parallel machines; memory fragmentation; memory map size; parallelism; partitionable SIMD/SPMD machines; partitionable interconnection networks; perfect memory maps; single program nature; user-controlled partitionable single-instruction-multiple data/single-program-multiple data; Broadcasting; Data engineering; High performance computing; Hypercubes; Large-scale systems; Multiprocessor interconnection networks; Oceans; Parallel machines; Parallel processing; Sufficient conditions;
fLanguage
English
Journal_Title
Parallel and Distributed Systems, IEEE Transactions on
Publisher
ieee
ISSN
1045-9219
Type
jour
DOI
10.1109/71.86105
Filename
86105
Link To Document