Adaptive Parallel Algorithm of Large Linear Equations
Zhao Hang-tao
Abstract
Zhao Hang-tao
Abstract
Researches on matrix type and corresponding Adaptive parallel algorithm of large linear system of equations.Proposes adaptive parallel algorithm,builds a cluster composed of 15 computers and also a parallel algorithms library based on a PETSC software engineering.Experiments based on adaptive parallel algorithm and the results show that the adaptive parallel algorithm can select the optimum algorithm from algorithm library and can compute the optimum nodes by the user’s preferences,so consequently to improve the universal property of parallel solver of linear system of equations and lower the difficulty of numerical analysis.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Researches on matrix type and corresponding Adaptive parallel algorithm of large linear system of equations.Proposes adaptive parallel algorithm,builds a cluster composed of 15 computers and also a parallel algorithms library based on a PETSC software engineering.Experiments based on adaptive parallel algorithm and the results show that the adaptive parallel algorithm can select the optimum algorithm from algorithm library and can compute the optimum nodes by the user’s preferences,so consequently to improve the universal property of parallel solver of linear system of equations and lower the difficulty of numerical analysis.
Key concepts: Computer science, Parallel algorithm, Solver, Algorithm, System of linear equations, Linear system, Linear equation, Analysis of parallel algorithms