DocumentCode :
652514
Title :
Multi-GPU Scalable Implementation of a Contour-Integral-Based Eigensolver for Real Symmetric Dense Generalized Eigenvalue Problems
Author :
Yano, Takao ; Futamura, Yasunori ; Sakurai, Takayasu
Author_Institution :
Dept. of Comput. Sci., Univ. of Tsukuba, Tsukuba, Japan
fYear :
2013
fDate :
28-30 Oct. 2013
Firstpage :
121
Lastpage :
127
Abstract :
We consider a parallel eigensolver for generalized eigenvalue problems for distributed GPU systems. In this paper, we propose a distributed parallel implementation of the Sakurai-Sugiura (SS) eigenvalue solver for solving generalized eigenvalue problems with real symmetric matrices using GPU linear algebra libraries. In the SS method, the target subspace is constructed from solutions of linear systems. The dominant part of this method is calculating solutions of linear equations. By assigning the solution of independent linear systems to each GPU, a coarse-grained parallelism can be obtained, and high scalability is expected. We also proposed the performance model of this implementation. We evaluate its parallel performance using numerical examples that involve medium-size dense matrices.
Keywords :
eigenvalues and eigenfunctions; graphics processing units; matrix algebra; parallel processing; GPU linear algebra libraries; Sakurai-Sugiura eigenvalue solver; contour-integral-based eigensolver; distributed GPU systems; distributed parallel implementation; linear equations; medium-size dense matrices; multi-GPU scalable implementation; parallel eigensolver; real symmetric dense generalized eigenvalue problems; real symmetric matrices; Eigenvalues and eigenfunctions; Graphics processing units; Libraries; Linear systems; Scalability; Sparse matrices; Symmetric matrices; contour integral; generalized eigenvalue problem; multi GPU;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
P2P, Parallel, Grid, Cloud and Internet Computing (3PGCIC), 2013 Eighth International Conference on
Conference_Location :
Compiegne
Type :
conf
DOI :
10.1109/3PGCIC.2013.24
Filename :
6681218
Link To Document :
بازگشت