Automatic alignment of array data and processes to reduce comminication time on DMPPs
Philippsen M (1995)
Publication Language: English
Publication Type: Conference contribution, Original article
Publication year: 1995
Publisher: ACM
City/Town: New York
Pages Range: 156-165
Conference Proceedings Title: Proceedings of the 5th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming (PPoPP'95)
Event location: Santa Barbara/CA/USA
ISBN: 0-89791-700-6
URI: http://www2.informatik.uni-erlangen.de/publication/download/Alignment-2.pdf
Abstract
The nested data-parallel programming model supports the design and implementation of irregular parallel algorithms. This paper describes work in progress to incorporate nested data parallelism into the object model of Java by developing a library of collection classes and adding a forall statement to the language. The collection classes provide parallel implementations of operations on the collections. The forall statement allows operations over the elements of a collection to be expressed in parallel. We distinguish between shape and data components in the collection classes, and use this distinction to simplify algorithm expression and to improve performance. We present initial performance data on two benchmarks with irregular algorithms, EM3d and Convex Hull, and on several microbenchmark programs.
Authors with CRIS profile
How to cite
APA:
Philippsen, M. (1995). Automatic alignment of array data and processes to reduce comminication time on DMPPs. In Wexelblat, Richard L. (Eds.), Proceedings of the 5th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming (PPoPP'95) (pp. 156-165). Santa Barbara/CA/USA, US: New York: ACM.
MLA:
Philippsen, Michael. "Automatic alignment of array data and processes to reduce comminication time on DMPPs." Proceedings of the Fifth ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming (PPoPP'95), Santa Barbara/CA/USA Ed. Wexelblat, Richard L., New York: ACM, 1995. 156-165.
BibTeX: Download