ParSy: Inspection and Transformation of Sparse Matrix Computations for Parallelism

ParSy: Inspection and Transformation of Sparse Matrix Computations for Parallelism
复制标题

DOI:
10.1109/sc.2018.00065
复制
发表时间:
2018-11
期刊:
SC18: International Conference for High Performance Computing, Networking, Storage and Analysis
影响因子:
--
通讯作者:
Kazem Cheshmi;Shoaib Kamil;M. Strout;M. Dehnavi
Kazem Cheshmi;Shoaib Kamil;M. Strout;M. Dehnavi
中科院分区:
其他
文献类型:
--
作者:
Kazem Cheshmi;Shoaib Kamil;M. Strout;M. Dehnavi

文献摘要

被引文献

相似文献

在这项工作中,我们描述ParSy,一个框架,使用一种新的检查策略沿着与一个简单的代码转换,以优化并行稀疏算法共享内存处理器。与现有的方法,可能会遭受负载不平衡和过度同步,ParSy使用一种新的任务粗化策略,以创建平衡良好的任务,可以并行执行,同时保持本地的内存访问。使用ParSy检查器和变换的代码性能优于现有的高度优化的稀疏矩阵算法,例如多核处理器上的Cholesky因式分解,与MKL Pardiso和PaStiX库相比,加速分别为2.8倍和3.1倍。
In this work, we describe ParSy, a framework that uses a novel inspection strategy along with a simple code transformation to optimize parallel sparse algorithms for shared memory processors. Unlike existing approaches that can suffer from load imbalance and excessive synchronization, ParSy uses a novel task coarsening strategy to create well-balanced tasks that can execute in parallel, while maintaining locality of memory accesses. Code using the ParSy inspector and transformation outperforms existing highly-optimized sparse matrix algorithms such as Cholesky factorization on multi-core processors with speedups of 2.8× and 3.1× over the MKL Pardiso and PaStiX libraries respectively.