A lightweight optimization selection method for Sparse Matrix-Vector Multiplication

TL;DR

This paper introduces a lightweight, architecture-aware method for optimizing sparse matrix-vector multiplication by selecting the best optimization based on matrix features or online profiling, improving performance across platforms.

Contribution

It presents two novel models for optimization selection in SpMV, enhancing performance stability and achieving significant speedups in multicore and manycore systems.

Findings

Achieves 29% performance gain on manycore platforms.

Effectively distinguishes and optimizes special matrices.

Maintains low runtime overhead suitable for iterative solvers.

Abstract

In this paper, we propose an optimization selection methodology for the ubiquitous sparse matrix-vector multiplication (SpMV) kernel. We propose two models that attempt to identify the major performance bottleneck of the kernel for every instance of the problem and then select an appropriate optimization to tackle it. Our first model requires online profiling of the input matrix in order to detect its most prevailing performance issue, while our second model only uses comprehensive structural features of the sparse matrix. Our method delivers high performance stability for SpMV across different platforms and sparse matrices, due to its application and architecture awareness. Our experimental results demonstrate that a) our approach is able to distinguish and appropriately optimize special matrices in multicore platforms that fall out of the standard class of memory bandwidth bound matrices, and b) lead to a significant performance gain of 29% in a manycore platform compared to an architecture-centric optimization, as a result of the successful selection of the appropriate optimization for the great majority of the matrices. With a runtime overhead equivalent to a couple dozen SpMV operations, our approach is practical for use in iterative numerical solvers of real-life applications.