scientific article; zbMATH DE number 3997592

From MaRDI portal
Publication:4724451





[For the entire collection see Zbl 0614.00022.] This paper describes several approaches to converting numerical algorithms for use on supercomputers. The machine architecture which will be considered here is a vector computer. These approaches will be illustrated by considering two algorithms which arise in modelling heat transfer in biomedical applications. One algorithm is a vectorized matrix-vector multiply, and the other is a new, vectorized, overlapping-block iterative method for solving finite element equations.











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4724451)