DocumentCode :
1051770
Title :
Region analysis: a parallel elimination method for data flow analysis
Author :
Lee, Yong-fong ; Ryder, Barbara G. ; Fiuczynski, Marc E.
Author_Institution :
Intel Corp., Santa Clara, CA, USA
Volume :
21
Issue :
11
fYear :
1995
fDate :
11/1/1995 12:00:00 AM
Firstpage :
913
Lastpage :
926
Abstract :
Parallel data flow analysis methods offer the promise of calculating detailed semantic information about a program at compile-time more efficiently than sequential techniques. Previous work on parallel elimination methods (Zobel, 1990) has been hampered by the lack of control over interval size; this can prohibit effective parallel execution of these methods. To overcome this problem, we have designed the region analysis method, a new elimination method for data flow analysis. Region analysis emphasizes flow graph partitioning to enable better load balancing in a more effective parallel algorithm. We present the design of region analysis and the empirical results we have obtained that indicate: the prevalence of large intervals in flow graphs derived from real programs; and the performance improvement of region analysis over parallel Allen-Cocke interval analysis. Our implementation analyzed programs from the Perfect Benchmarks and netlib running on a Sequent Symmetry S81
Keywords :
data flow analysis; data flow graphs; parallel algorithms; parallel programming; program compilers; resource allocation; software performance evaluation; Perfect Benchmarks; Sequent Symmetry S81; compile-time; flow graph partitioning; interval analysis; interval size; load balancing; netlib; parallel algorithm; parallel data flow analysis methods; parallel elimination method; parallel execution; performance improvement; program optimization; region analysis; semantic information; Algorithm design and analysis; Computer science; Data analysis; Flow graphs; Information analysis; Load flow analysis; Parallel algorithms; Performance analysis; Size control; Software tools;
fLanguage :
English
Journal_Title :
Software Engineering, IEEE Transactions on
Publisher :
ieee
ISSN :
0098-5589
Type :
jour
DOI :
10.1109/32.473220
Filename :
473220
Link To Document :
بازگشت