Parallelization for Multi-dimensional Loops with Non-uniform Dependences.

Saved in:

Bibliographic Details
Title:	Parallelization for Multi-dimensional Loops with Non-uniform Dependences.
Authors:	Sam-Jin Jeong¹ sjjeong@bu.ac.kr
Source:	IAENG International Journal of Computer Science. Feb2018, Vol. 45 Issue 1, p107-111. 5p.
Subjects:	Loops (Group theory), Parallelizing compilers, Software engineers, Integers, Parallel algorithms
Abstract:	This paper presents two enhanced partitioning algorithms in order to improve parallelization of multidimensional loops with flow and anti-dependences Using distance of the first dependence, we show the first general enhanced algorithm of single-loops with both flow and anti-dependences. Using the first algorithm, loop interchanging method, and cycle shrinking method, we present parallelization of nested loops with simple subscripts. Using the second algorithm, we also present parallelization to multi-dimensional loops with both flow and anti-dependences. Our two presented algorithms show enhanced loop parallelization of multi-dimensional loops with both dependences. We will improve our proposed algorithms for multi-dimensional space with multiple dependences. [ABSTRACT FROM AUTHOR]
	Copyright of IAENG International Journal of Computer Science is the property of International Association of Engineers (IAENG) and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database:	Engineering Source

Description
Abstract:	This paper presents two enhanced partitioning algorithms in order to improve parallelization of multidimensional loops with flow and anti-dependences Using distance of the first dependence, we show the first general enhanced algorithm of single-loops with both flow and anti-dependences. Using the first algorithm, loop interchanging method, and cycle shrinking method, we present parallelization of nested loops with simple subscripts. Using the second algorithm, we also present parallelization to multi-dimensional loops with both flow and anti-dependences. Our two presented algorithms show enhanced loop parallelization of multi-dimensional loops with both dependences. We will improve our proposed algorithms for multi-dimensional space with multiple dependences. [ABSTRACT FROM AUTHOR]
ISSN:	1819656X