Paper Info
Title
A study on popular auto-parallelization frameworks.
Abstract
We study five popular auto-parallelization frameworks (Cetus, Par4all, Rose, ICC, and Pluto) and compare them qualitatively as well as quantitatively. All the frameworks primarily deal with loop parallelization but differ in the techniques used to identify parallelization opportunities. Due to this variance, various aspects, such as certain loop transformations, are supported only in a few frameworks. The frameworks exhibit varying abilities in handling loop-carried dependence and, therefore, achieve different amounts of speedup on widely used PolyBench and NAS parallel benchmarks. In particular, Intel C Compiler (ICC) fares as an overall good parallelizer. Our study also highlights the need for more sophisticated analyses, user-driven parallelization, and meta-auto-parallelizer that provides combined benefits of various frameworks.
Year
DOI
Venue
2019
10.1002/cpe.5168
CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE
Keywords
DocType
Volume
loop-carried dependence,loop transformations,privatization,vectorization
Journal
31.0
Issue
ISSN
Citations 
17.0
1532-0626
0
PageRank 
References 
Authors
0.34
0
4
Name
Order
Citations
PageRank
S. Prema100.34
Rupesh Nasre234121.02
R. Jehadeesan300.34
B. K. Panigrahi400.34