A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2005; you can also visit the original URL.
The file type is application/pdf.
Processor reordering algorithms toward efficient GEN_BLOCK redistribution
2001
Proceedings of the 2001 ACM symposium on Applied computing - SAC '01
Preserved Fulltext
The use of data redistribution represents a performance tradeoff between the expected higher efficiency of a new distribution for subsequent computation and the communication cost of redistributing the data among processor memories. This paper focuses on reducing the communication cost in GEN_BLOCK redistribution using a logical processor reordering method. We propose three algorithms; Number_Oriented, Size_Oriented and Numsize_Oriented. According to experiments on CRAY T3E, the algorithms show
doi:10.1145/372202.372457
dblp:conf/sac/LeeYKP01
fatcat:xsqxtlemtvcgbgsyftiugefl7u