TRENDS IN SYSTOLIC AND CELLULAR COMPUTATION

[摘要] A profile is given of current research, as it pertains to computational mathematics, on Very Large Scale Integration (VLSI) array processors. In this type of parallel computers, the cells of the array operate in Single Instruction Multiple Data (SIMD) mode and algorithms are executed in systolic or cellular fashion. The focus of the presentation is on linear algebraic techniques. A systolization process is illustrated by matching an arbitrary gaxpy operation onto a fixed-size square array processor. Two recent systolic methods, for solving systems of linear equations, one iterative and the other direct, are described. A cellular algorithm for a fast Fourier transform, based on a new implementation on rectangular array processors of the perfect shuffle permutation, is then derived. Using annotated lists of recent references, snapshots of active research areas are given on systolic linear solvers, the singular value decomposition, artificial neural networks, and the simulated annealing algorithm.

[发布日期] 1991-12-23

[效力级别]  Proceedings Paper

[关键词] PARALLEL COMPUTATIONAL LINEAR ALGEBRA;MESH ARRAY PROCESSORS;SYSTOLIC AND CELLULAR ALGORITHMS