Efficient Algorithm Design of Optimizing SpMV on GPU.
Genshen Chu,Yuanjie He,Lingyu Dong,Zhezhao Ding,Dandan Chen,He Bai,Xuesong Wang,Changjun Hu
DOI: https://doi.org/10.1145/3588195.3593002
2023-01-01
Abstract:Sparse matrix-vector multiplication (SpMV) is a fundamental building block for various numerical computing applications. However, most existing GPU-SpMV approaches may suffer from either long preprocessing overhead, load imbalance, format conversion, bad memory access patterns. In this paper, we proposed two new SpMV algorithms:flat andline-enhance, as well as their implementations, for GPU systems to overcome the above shortcomings. Our algorithms work directly on the CSR sparse matrix format. To achieve high performance: 1) for load balance, theflat algorithm uses non-zero splitting andline-enhance uses a mix of row and non-zero splitting; 2) memory access patterns are designed for both algorithms for data loading, storing and reduction steps; and 3) an adaptive approach is proposed to select appropriate algorithm and parameters based on matrix characteristics. We evaluate our methods using theSuiteSparse Matrix Collection on AMD and NVIDIA GPU platforms. Average performance improvements of 424%, 741%, 49%, 46%, 72% are achieved when comparing our adaptive approach with CSR-Vector, CSR-Adaptive, HOLA, cuSparse and merge-based SpMV, respectively. In bandwidth tests, our approach can also achieve a high memory bandwidth, which is very close to the peak memory bandwidth.