DocumentCode :
2061454
Title :
PoLABACK: parallel factorization algorithms with algorithmic blocking
Author :
Choi, Jaeyoung
Author_Institution :
Sch. of Comput., Soongsil Univ., Seoul, South Korea
Volume :
1
fYear :
2000
fDate :
14-17 May 2000
Firstpage :
200
Abstract :
Since parallel computers have different performance ratios of computation and communication, the optimal computational block sizes are different from one another to generate the maximum performance of an algorithm. Too small or large a block size makes getting good performance on a machine nearly impossible. In such a case, getting a better performance may require a complete redistribution of the data matrix. We present PoLAPACK factorization routines, including LU, QR, and Cholesky factorizations, with an "algorithmic blocking" on 2-dimensional block cyclic data distribution. With the algorithmic blocking, it is possible to obtain the near optimal performance irrespective of the physical block size. The routines are implemented on the SGI/Cray T3E and compared with the corresponding ScaLAPACK factorization routines.
Keywords :
distributed memory systems; matrix algebra; parallel algorithms; Cholesky factorizations; LU factorisation; PoLABACK; QR factorisation; SGI/Cray T3E; ScaLAPACK factorization; algorithmic blocking; data matrix; linear algebra algorithms; optimal computational block sizes; parallel computers; parallel factorization algorithms; performance ratios;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
High Performance Computing in the Asia-Pacific Region, 2000. Proceedings. The Fourth International Conference/Exhibition on
Conference_Location :
Beijing, China
Print_ISBN :
0-7695-0589-2
Type :
conf
DOI :
10.1109/HPC.2000.846546
Filename :
846546
Link To Document :
بازگشت