Dong Chun-Li, National Digital Switching System Engineering & Technology Research Center, China
Zhao Rong-Cai, National Digital Switching System Engineering & Technology Research Center, China
Ma Zhuo-Jie, National Digital Switching System Engineering & Technology Research Center, China
Li Dong-Hong, National Digital Switching System Engineering & Technology Research Center, China
Minimizing communication by increasing the locality of data references is an important optimization for achieving high performance on distributed memory machines. But in the progress of decomposition, reorganization is inevitable. And the communication produced by reorganization is inevitable too. In this paper, the authors present a linear decomposition algorithm that automatically finding computation and data decomposition, including finding data and computations decomposition that has data reorganization communication. And the authors improve the method and reduce the communication cost by merging parallel regions with the same data decomposition. Keywords: Parallelizing compiler, MPI, data reorganization, parallel region merging, data decomposition, computation decomposition
Citation:
Dong Chun-Li, Zhao Rong-Cai, Ma Zhuo-Jie, Li Dong-Hong, "Communication Optimization on Automatic Program Parallelization," snpd, vol. 1, pp.437-442, Eighth ACIS International Conference on Software Engineering, Artificial Intelligence, Networking, and Parallel/Distributed Computing (SNPD 2007), 2007
PURCHASE ARTICLE: $19
IEEE Xplore Subscribers
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb