Physics

Out-of-core algorithm

Supercomputer

Operating system

Algorithm

Matrix (chemical analysis)

Matrix multiplication

Parallel computing

Composite material

Exploit

Materials science

Overhead (engineering)

Computer science

Quantum mechanics

Multi-core processor

Computer security

Quantum

ACM Transactions on Architecture and Code Optimization

We present a systematic methodology for optimizing batched matrix multiplications on SW26010 many-core processor of the Sunway TaihuLight supercomputer. Five surrogate algorithms and a machine learning–based algorithm selector are proposed to fully exploit the computing capability of SW26010 and cope with the sophisticated algorithm characteristics of batched matrix multiplications. Experiment results show that the algorithm selector is able to adaptively choose the appropriate algorithm for various matrix shapes and batch sizes with low overhead and high accuracy. In particular, the optimized batched matrix multiplications can substantially outperform the non-batched version and reach around 84.8% of the performance upper bound.

Enabling Highly Efficient Batched Matrix Multiplications on SW26010 Many-core Processor