Combinatorics

Physics

Aerospace engineering

Multiplication (music)

Mathematics

Algorithm

Matrix (chemical analysis)

Engineering

Gaussian

Matrix multiplication

Parallel computing

Composite material

Reuse

Materials science

Range (aeronautics)

Computer science

Sparse matrix

Quantum mechanics

Dataflow

Waste management

Quantum

Sparse matrix-sparse matrix multiplication (spMspM) is at the heart of a wide range of scientific and machine learning applications. spMspM is inefficient on general-purpose architectures, making accelerators attractive. However, prior spMspM accelerators use inner- or outer-product dataflows that suffer poor input or output reuse, leading to high traffic and poor performance. These prior accelerators have not explored Gustavson's algorithm, an alternative spMspM dataflow that does not suffer from these problems but features irregular memory access patterns that prior accelerators do not support.

Gamma: leveraging Gustavson’s algorithm to accelerate sparse matrix multiplication