期刊
PARALLEL COMPUTING
卷 101, 期 -, 页码 -出版社
ELSEVIER
DOI: 10.1016/j.parco.2020.102724
关键词
Sparse approximate inverse; Preconditioning; CUDA; GPU
资金
- Natural Science Foundation of China [61872422]
- Natural Science Foundation of Zhejiang Province, China [LY19F020028]
- Natural Science Foundation of Jiangsu Province, China [BK20171480]
This study introduces an efficient sparse approximate inverse preconditioning algorithm, GSPAI-Adaptive, on multiple GPUs. It presents a thread-adaptive allocation strategy for constructing the preconditioner and computes each component of the preconditioner in parallel inside a thread group of GPU, showing advantages over popular preconditioning algorithms and a latest parallel sparse approximate inverse preconditioning algorithm in experimental results.
In this study, we present an efficient thread-adaptive sparse approximate inverse preconditioning algorithm on multiple GPUs, called GSPAI-Adaptive. For our proposed GSPAI-Adaptive, there are the following novelties: (1) a thread-adaptive allocation strategy is presented for each column of the preconditioner, and (2) a parallel framework of constructing the sparse approximate inverse preconditioner is proposed on multiple GPUs, and (3) each component of the preconditioner is computed in parallel inside a thread group of GPU. Experimental results show that GSPAI-Adaptive is effective, and is advantageous over the popular preconditioning algorithms in two public libraries, and a latest parallel sparse approximate inverse preconditioning algorithm.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据