TY - GEN
T1 - From sparse matrix to optimal GPU CUDA sparse matrix vector product implementation
AU - El Zein, Ahmed H.
AU - Rendell, Alistair P.
PY - 2010
Y1 - 2010
N2 - The CUDA model for GPUs presents the programmer with a plethora of different programming options. These includes different memory types, different memory access methods, and different data types. Identifying which options to use and when is a non-trivial exercise. This paper explores the effect of these different options on the performance of a routine that evaluates sparse matrix vector products. A process for analysing performance and selecting the subset of implementations that perform best is proposed. The potential for mapping sparse matrix attributes to optimal CUDA sparse matrix vector product implementation is discussed.
AB - The CUDA model for GPUs presents the programmer with a plethora of different programming options. These includes different memory types, different memory access methods, and different data types. Identifying which options to use and when is a non-trivial exercise. This paper explores the effect of these different options on the performance of a routine that evaluates sparse matrix vector products. A process for analysing performance and selecting the subset of implementations that perform best is proposed. The potential for mapping sparse matrix attributes to optimal CUDA sparse matrix vector product implementation is discussed.
UR - http://www.scopus.com/inward/record.url?scp=77954926909&partnerID=8YFLogxK
U2 - 10.1109/CCGRID.2010.81
DO - 10.1109/CCGRID.2010.81
M3 - Conference contribution
SN - 9781424469871
T3 - CCGrid 2010 - 10th IEEE/ACM International Conference on Cluster, Cloud, and Grid Computing
SP - 808
EP - 813
BT - CCGrid 2010 - 10th IEEE/ACM International Conference on Cluster, Cloud, and Grid Computing
T2 - 10th IEEE/ACM International Symposium on Cluster, Cloud, and Grid Computing, CCGrid 2010
Y2 - 17 May 2010 through 20 May 2010
ER -