Title: Programming paradigms for parallel finite element analysis and structural optimization
Abstract: We live in an era where many personal workstations use multi-core processors where manufacturers resort to adding cores onto the chip in a bid to increase computing power. Clusters of multi-core computers are now common in industry and research labs. However, the clock frequency of commodity processors is unlikely to surpass 4 GHz anytime soon, and so users/programmers need to study the area of parallel computing and modify the software they create in order to take advantage of the performance enhancement available due to the added cores (multi-core)/processors (cluster). Computing tools do exist to enable modifications to programs without a complete overhaul, for example programming with parallel MATLAB/SciLAB, High Performance Fortran, OpenMP for shared memory SYSTEMS, MPI for distributed memory systems, are well established. The goal of the present paper is to introduce the concepts behind parallel programming and introduce programming paradigms for FEA and FEA-based optimization in the simplest possible manner, so as to allow computational users in the area of structural mechanics to effectively benefit from the increase in available computing power.
Publication Year: 2010
Publication Date: 2010-05-16
Language: en
Type: preprint
Access and Citation
Cited By Count: 1
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot