Sciweavers

Share
IPPS
1998
IEEE

A Generalized Framework for Global Communication Optimization

11 years 9 months ago
A Generalized Framework for Global Communication Optimization
In distributed-memory message-passing architectures reducing communication cost is extremely important. In this paper, we present a technique to optimize communication globally. Our approach is based on a combination of linear algebra framework and dataflow analysis, and can take arbitrary control flow into account. The distinctive features of the algorithm are its accuracy in keeping communication set information and its support for general alignments and distributions including block-cyclic distributions. The method is currently being implemented in the PARADIGM compiler. The preliminary results show that the technique is effective in reducing both number as well as volume of the communication.
Mahmut T. Kandemir, Prithviraj Banerjee, Alok N. C
Added 05 Aug 2010
Updated 05 Aug 2010
Type Conference
Year 1998
Where IPPS
Authors Mahmut T. Kandemir, Prithviraj Banerjee, Alok N. Choudhary, J. Ramanujam, U. Nagaraj Shenoy
Comments (0)
books