Developments of iterative-type preconditioning techniques for large linear systems and their application

2015 Fiscal Year Final Research Report

• Project/Area Number: 25870099
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Computational science; High performance computing
• Research Institution: University of Tsukuba
• Principal Investigator: Imakura Akira 筑波大学, システム情報系, 助教 (60610045)
• Project Period (FY): 2013-04-01 – 2016-03-31
• Keywords: 高性能計算 / 数値解析 / 大規模連立一次方程式 / 反復法 / 前処理技術

Outline of Final Research Achievements

In this project, we aim to develop efficient algorithms for solving large linear systems and apply to some real applications. In order to achieve these purposes, we developed iterative-type preconditioning techniques which have high speed, high stability and high parallel efficiency. We also provided parameters optimization techniques for improving convenience of the proposed preconditioning techniques.
The proposed preconditioning techniques and their parameter optimization techniques were applied to supernova simulation. The proposed techniques made it possible to solve efficiently difficult problems for traditional techniques. Then, the proposed techniques adopted to a supernova simulation code and are beginning to make achievements.

Free Research Field

数値解析