Affiliation:
1. University of Passau, Germany
Abstract
Performance optimization of stencil codes requires data locality improvements. The polyhedron model for loop transformation is well suited for such optimizations with established techniques, such as the PLuTo algorithm and diamond tiling. However, in the domain of our project ExaStencils, stencil codes, it fails to yield optimal results. As an alternative, we propose a new, optimized, multi-dimensional polyhedral search space exploration and demonstrate its effectiveness: we obtain better results than existing approaches in several cases. We also propose how to specialize the search for the domain of stencil codes, which dramatically reduces the exploration effort without significantly impairing performance.
Funder
German Research Foundation
Publisher
Association for Computing Machinery (ACM)
Subject
Hardware and Architecture,Information Systems,Software
Cited by
5 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
1. EPSILOD: efficient parallel skeleton for generic iterative stencil computations in distributed GPUs;The Journal of Supercomputing;2023-01-14
2. Scalable Distributed High-Order Stencil Computations;SC22: International Conference for High Performance Computing, Networking, Storage and Analysis;2022-11
3. ExaStencils: Advanced Multigrid Solver Generation;Software for Exascale Computing - SPPEXA 2016-2019;2020
4. Automatic Differentiation for Adjoint Stencil Loops;Proceedings of the 48th International Conference on Parallel Processing;2019-08-05
5. Speeding up Iterative Polyhedral Schedule Optimization with Surrogate Performance Models;ACM Transactions on Architecture and Code Optimization;2019-01-08