Hyper-Systolic Matrix Multiplication
Thomas Lippert, Nikolay Petkov, Paolo Palazzari, and Klaus Schilling
TL;DR
This paper introduces a new parallel matrix multiplication algorithm called hyper-systolic, which efficiently utilizes a one-dimensional processor model and is adaptable to various parallel computing systems for different matrix operations.
Contribution
The paper presents the hyper-systolic algorithm, a novel parallel approach for matrix multiplication that is flexible and applicable to multiple types of matrix computations on diverse parallel architectures.
Findings
Applicable to all types of parallel systems
Handles matrix-vector and transposed matrix products
Offers a new approach to parallel matrix multiplication
Abstract
A novel parallel algorithm for matrix multiplication is presented. The hyper-systolic algorithm makes use of a one-dimensional processor abstraction. The procedure can be implemented on all types of parallel systems. It can handle matrix-vector multiplications as well as transposed matrix products.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.