Some parallel numerical methods in solving partial differential equations

• Main Authors: Alias, Norma, Satam, Noriza, Darwis, Roziha, Hamzah, Norhafizah, A. Ghaffar, Zarith Safiza, Islam, Md. Rajibul
• Format: Conference or Workshop Item
• Language: English
• Published: 2010
• Subjects: QA75 Electronic computers. Computer science; QA Mathematics
• Online Access: http://eprints.utm.my/id/eprint/9840/1/V2-G387.pdf; http://eprints.utm.my/id/eprint/9840/; http://dms.library.utm.my:8080/vital/access/manager/Repository/vital:101284

This paper will discuss the solution of twodimensional partial differential equations (PDEs) using some parallel numerical methods namely Gauss Seidel and Red Black Gauss Seidel. The selected two-dimensional PDE to solve in this paper are of parabolic and elliptic type. Parallel Virtual Machine (PVM) is used in support of the communication among all microprocessors of Parallel Computing System. PVM is well known as a software system that enables a collection of heterogeneous computers to be used as coherent and flexible concurrent computational resource. The numerical results will be presented graphically and parallel performance measurement by Gauss Seidel and Red Gauss Seidel methods will be evaluated in terms of execution time, speedup, efficiency, effectiveness and temporal performance. Performance evaluations are critical as this paper aimed to fabricate an efficient Two-Dimensional PDE Solver (TDPDES). This new well-organized TDPDES technique will enhance the research and analysis procedure of many engineering and mathematic fields.