MPI Reduction Operations for Sparse Floating-point Data
MPI Reduction Operations for Sparse Floating-point Data
复制标题
稀疏浮点数据的 MPI 约简运算
DOI：
--
复制
发表时间：
2008
期刊：
影响因子：
--
通讯作者：
G. Rünger
中科院分区：
文献类型：
--
作者：
Michael Hofmann;G. Rünger
本文提出了一种MPI_Reduce的流水线算法，该算法使用游程编码（RLE）方案来提高稀疏浮点数据的全局约简。RLE方案被直接纳入到还原过程中，并且在最坏的情况下仅导致低开销。RLE方案的高吞吐量也允许在使用高性能互连时提高性能。随机样本数据和稀疏向量数据从并行有限元应用程序被用来证明新的约简算法的性能与InfiniBand互连的HPC集群。
This paper presents a pipeline algorithm for MPI_Reduce that uses a Run Length Encoding(RLE) scheme to improve the global reduction of sparse floating-point data. The RLE scheme is directly incorporated into the reduction process and causes only low overheads in the worst case. The high throughput of the RLE scheme allows performance improvements when using high performance interconnects, too. Random sample data and sparse vector data from a parallel FEM application is used to demonstrate the performance of the new reduction algorithm for an HPC Cluster with InfiniBand interconnects.