Hierarchical Parallel Matrix Multiplication on Large-Scale Distributed Memory Platforms

TL;DR

This paper presents HSUMMA, a hierarchical parallel matrix multiplication algorithm that significantly reduces communication costs on large-scale distributed memory platforms, outperforming SUMMA especially at higher core counts.

Contribution

The paper introduces HSUMMA, a novel hierarchical redesign of SUMMA, to lower communication costs in parallel matrix multiplication on large-scale systems.

Findings

Up to 2.08x reduction in communication cost on 2048 cores.

Up to 5.89x reduction on 16384 cores.

Effective scalability demonstrated on IBM BlueGene-P.

Abstract

Matrix multiplication is a very important computation kernel both in its own right as a building block of many scientific applications and as a popular representative for other scientific applications. Cannon algorithm which dates back to 1969 was the first efficient algorithm for parallel matrix multiplication providing theoretically optimal communication cost. However this algorithm requires a square number of processors. In the mid 1990s, the SUMMA algorithm was introduced. SUMMA overcomes the shortcomings of Cannon algorithm as it can be used on a non-square number of processors as well. Since then the number of processors in HPC platforms has increased by two orders of magnitude making the contribution of communication in the overall execution time more significant. Therefore, the state of the art parallel matrix multiplication algorithms should be revisited to reduce the communication cost further. This paper introduces a new parallel matrix multiplication algorithm, Hierarchical SUMMA (HSUMMA), which is a redesign of SUMMA. Our algorithm reduces the communication cost of SUMMA by introducing a two-level virtual hierarchy into the two-dimensional arrangement of processors. Experiments on an IBM BlueGene-P demonstrate the reduction of communication cost up to 2.08 times on 2048 cores and up to 5.89 times on 16384 cores.