Litcius/Paper detail

HARP: Hardware-Based Pseudo-Tiling for Sparse Matrix Multiplication Accelerator

JinKwon Kim, Myeongjae Jang, H.K. Nam, Soontae Kim

202312 citationsDOIOpen Access PDF

Abstract

General sparse matrix-matrix multiplication (SpGEMM) is a memory-bound workload, due to the compression format used. To minimize data movements for input matrices, outer product accelerators have been proposed. Since these accelerators access input matrices only once and then generate numerous partial products, managing the generated partial products is the key optimization factor. To reduce the number of partial products handled, the state-of-the-art accelerator uses software to tile an input matrix. However, the software-based tiling has three limitations. First, a user manually executes the tiling software and manages the tiles. Second, generating a compression format for each tile incurs memory-intensive operations. Third, an accelerator that uses the compression format cannot skip ineffectual accesses for input matrices.

Topics & Concepts

Computer scienceMatrix multiplicationMultiplication (music)Parallel computingSoftwareSparse matrixMatrix (chemical analysis)Hardware accelerationTileComputer hardwareComputational scienceOperating systemMathematicsComposite materialCombinatoricsArtQuantumMaterials sciencePhysicsGaussianVisual artsQuantum mechanicsParallel Computing and Optimization TechniquesDistributed and Parallel Computing SystemsAdvanced Data Storage Technologies