In this paper we study the performance of constraint-basedlocalsearch solvers on a GPU. The massively parallel architecture of the GPU makes it possible to explore parallelism at two different levels inside the loca...
详细信息
ISBN:
(纸本)9781479927289
In this paper we study the performance of constraint-basedlocalsearch solvers on a GPU. The massively parallel architecture of the GPU makes it possible to explore parallelism at two different levels inside the localsearch algorithm. First, by executing multiple copies of the algorithm in a multiwalk manner and, second, by evaluating large neighborhoods in parallel in a single-walk manner. Experiments on three well-known problem benchmarks indicate that the current GPU implementation is up to 17 times faster than a well-tuned sequential algorithm implemented on a desktop computer.
暂无评论