Presentation | 2015-01-30 Implementation of Sparse Matrix-Vector Multiplication on GPU and Its Application to the Conjugate Gradient Method Shotaro ASANO, Masato INAGI, Shinobu NAGAYAMA, Shin'ichi WAKABAYASHI, |
---|---|
PDF Download Page | ![]() |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Numerical simulations are offten performed by converting complex partial differential equations into a system of discrete linear equations, and solving the system of linear equations. Therefore, the speed of numerical simulations depends on how fast a system of linear equations is solved. In this research, thus, we focus on the conjugate gradient method that is a solver of a system of linear equations, and experimentally evaluate its parallel implementation on a GPU to investigate how fast it is. Our previous research showed that matrix-vector products account for a large portion of the total computation time of the conjugate gradient method. In addition, matrices for a system of linear equations used for numerical simulations are usually very sparse. To compute the matrix-vector products efficiently, various representations for sparse matrices, such as COO and CSR, are used. In this paper, we implement matrix-vector products on a GPU and a CPU (in both single- and multi-thread implementations) using those representations, and compare their computation time to find out a representation appropriate for their implementations. In this paper, we also quantitatively evaluate how faster the conjugate gradient method is than the existing one when the appropriate representation of sparse matrices is used. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Simultaneous linear equations / Conjugate gradient method / Sparse matrix / GPU |
Paper # | VLD2014-141,CPSY2014-150,RECONF2014-74 |
Date of Issue |
Conference Information | |
Committee | RECONF |
---|---|
Conference Date | 2015/1/22(1days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | |
Vice Chair | |
Secretary | |
Assistant |
Paper Information | |
Registration To | Reconfigurable Systems (RECONF) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Implementation of Sparse Matrix-Vector Multiplication on GPU and Its Application to the Conjugate Gradient Method |
Sub Title (in English) | |
Keyword(1) | Simultaneous linear equations |
Keyword(2) | Conjugate gradient method |
Keyword(3) | Sparse matrix |
Keyword(4) | GPU |
1st Author's Name | Shotaro ASANO |
1st Author's Affiliation | Graduate School of Information Sciences, Hiroshima City University() |
2nd Author's Name | Masato INAGI |
2nd Author's Affiliation | Graduate School of Information Sciences, Hiroshima City University |
3rd Author's Name | Shinobu NAGAYAMA |
3rd Author's Affiliation | Graduate School of Information Sciences, Hiroshima City University |
4th Author's Name | Shin'ichi WAKABAYASHI |
4th Author's Affiliation | Graduate School of Information Sciences, Hiroshima City University |
Date | 2015-01-30 |
Paper # | VLD2014-141,CPSY2014-150,RECONF2014-74 |
Volume (vol) | vol.114 |
Number (no) | 428 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |