English

s-Step Orthomin and GMRES implemented on parallel computers

Numerical Analysis 2020-01-28 v2 Distributed, Parallel, and Cluster Computing Numerical Analysis

Abstract

The Orthomin ( Omin ) and the Generalized Minimal Residual method ( GMRES ) are commonly used iterative methods for approximating the solution of non-symmetric linear systems. The s-step generalizations of these methods enhance their data locality parallel and properties by forming s simultaneous search direction vectors. Good data locality is the key in achieving near peak rates on memory hierarchical supercomputers. The theoretical derivation of the s-step Arnoldi and Omin has been published in the past. Here we derive the s-step GMRES method. We then implement s-step Omin and GMRES on a Cray-2 hierarchical memory supercomputer.

Cite

@article{arxiv.2001.04886,
  title  = {s-Step Orthomin and GMRES implemented on parallel computers},
  author = {A. T. Chronopoulos and S. K. Kim},
  journal= {arXiv preprint arXiv:2001.04886},
  year   = {2020}
}
R2 v1 2026-06-23T13:11:01.281Z