Performance of Linear Algebra Codes on the CRAY-1

Publisher Website
Google Scholar
Abstract
As of 1978, the CRAY-1 has potentially the highest floating-point computation rate in both scalar and vector modes of any available general scientific processor. Exploitation of this superior capability is the subject of this paper. In particular, we consider (1) mathematical models, software issues, and performance measures of vector machines, (2) algorithmic and coding procedures related to development of linear algebra codes on the CRAY-1, and (3) performance of direct solution codes commonly encountered in reservoir simulation.

This publication has 0 references indexed in Scilit: