Abstract
A parallel nonlinear dynamic finite element analysis code (PANDA) which can predict the response of three-dimensional shell structures has been developed for execution on a hypercube computer. This code is implemented in conjunction with a previously developed mesh decomposition algorithm. The performance of the parallel code has been tested on a 32-processor Intel iPSC/d5 hypercube. Performance results have been obtained for a range of problems which exhibit large deformation, elastic-plastic dynamic response, including an idealized automotive lower side-rail subject to impact loading. Speedup factors of greater than 31 have been achieved when executing on 32 processors. Efficiencies of 99% have been obtained by assigning as few as eight shell elements to each processor. It is demonstrated that these high efficiencies decrease very slowly as the number of processors is increased, provided the size of the problem is appropriately scaled. Since complex finite element models may contain 10,000 or more elements, this demonstrates that massively parallel hypercube machines have enormous potential for solving these types of structural analysis problems.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.