期刊
JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING
卷 144, 期 -, 页码 55-67出版社
ACADEMIC PRESS INC ELSEVIER SCIENCE
DOI: 10.1016/j.jpdc.2020.05.021
关键词
Graphics processing unit; GPU computing; Quadratic programming; Alternating direction method of multipliers
资金
- European Research Council (ERC) under the European Union [787845]
The alternating direction method of multipliers (ADMM) is a powerful operator splitting technique for solving structured convex optimization problems. Due to its relatively low per-iteration computational cost and ability to exploit sparsity in the problem data, it is particularly suitable for large-scale optimization. However, the method may still take prohibitively long to compute solutions to very large problem instances. Although ADMM is known to be parallelizable, this feature is rarely exploited in real implementations. In this paper we exploit the parallel computing architecture of a graphics processing unit (GPU) to accelerate ADMM. We build our solver on top of OSQP, a state-of-the-art implementation of ADMM for quadratic programming. Our open-source CUDA C implementation has been tested on many large-scale problems and was shown to be up to two orders of magnitude faster than the CPU implementation. (C) 2020 Elsevier B.V. All rights reserved.
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