A Systematic Survey of General Sparse Matrix-matrix Multiplication

General Sparse Matrix-Matrix Multiplication (SpGEMM) has attracted much attention from researchers in graph analyzing, scientific computing, and deep learning. Many optimization techniques have been developed for different applications computing architectures over the past decades. The objective of this paper is to provide a structured comprehensive overview researches on SpGEMM. Existing grouped into categories based target design choices. Covered topics include typical applications, compression formats, general formulations, key problems techniques, architecture-oriented optimizations, programming models. rationales algorithms are analyzed summarized. This survey sufficiently reveals latest progress SpGEMM research 2021. Moreover, thorough performance comparison existing implementations presented. Based our findings, we highlight future directions, which encourage better later studies.