On the two-stage multisplitting iteration methods for linear complementarity problems

In this work, the two-stage multisplitting iteration methods based on the equivalent modulus equations are analyzed for solving linear complementarity problems. New convergence results are presented where the convergence domains of the parameter matrices are enlarged compared the existing literature...

Full description

Saved in:

Bibliographic Details
Published in:	Applied mathematics and computation Vol. 475; p. 128741
Main Authors:	Guo, Wenxiu, Zheng, Hua, Lu, Xiaoping, Zhang, Yongxiong
Format:	Journal Article
Language:	English
Published:	Elsevier Inc 15.08.2024
Subjects:	Linear complementarity problem Modulus-based multisplitting OpenACC Parallel computation Two-stage method 65F10 90C33 Two-stage method Parallel computation OpenACC Linear complementarity problem Modulus-based multisplitting 65Y05
ISSN:	0096-3003
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	In this work, the two-stage multisplitting iteration methods based on the equivalent modulus equations are analyzed for solving linear complementarity problems. New convergence results are presented where the convergence domains of the parameter matrices are enlarged compared the existing literatures. Furthermore, the improved domains are validated by numerical examples with parallel computations using OpenACC. •Propose two convergence theorems of the two-stage multisplitting iteration methods for LCPs with new proof techniques.•The given results can improve the domain of the parameter matrices in the existing literatures.•Design an OpenACC framework where the efficiency of the proposed methods is shown to be improved with good speedups.
ISSN:	0096-3003
DOI:	10.1016/j.amc.2024.128741