%0 Journal Article
%A DAI Zhi-tao
%A HUANG Zhi-bin
%A TANG Yang
%A ZHAO Da-fei
%T High Performance Row-Based Hashing GPU SpGEMM
%D 2019
%R 10.13190/j.jbupt.2018-252
%J Journal of Beijing University of Posts and Telecommunications
%P 106-113
%V 42
%N 3
%X Aiming at the performance problem of general sparse matrix-matrix multiplication (SpGEMM), a graphics processing unit (GPU)-accelerate SpGEMM algorithm based on task classification and low-latency Hashing table, RBSPARSE, was presented in the paper. RBSPARSE consists of a low-cost pre-analysis method to identify the complexity of sub-tasks, and a Hashing table-based algorithm which could utilize low-latency shared memory to achieve max efficiency. By taking the load balancing issue and the memory latency issue into consideration, RBSPARSE could significantly reduce the overall time in computation. RBSparse and BHSparse are compared. BHSparse is the previous state-of-the-art algorithm for SpGEMM. The result shows that our algorithm is 3.1 times faster than BHSparse on average, and could achieve a maximum 14.49 times faster speed in the best scenario.
%U https://journal.bupt.edu.cn/EN/10.13190/j.jbupt.2018-252