Abstract
Parallel structural analysis of framed structures on shared memory multiprocessor computers is presented. Emphasis is directed towards parallelizing each computational step of the solution process. Parallel algorithms are developed to maintain the workload balance throughout the solution process and to assure best concurrent performance and speed-up. A substructuring algorithm is used in order to achieve the workload balance during the solution of the generalized displacements. Concurrency is achieved through the use of the notion of cheap concurrency and the concept of threads. Two strategies for avoiding a racing condition are presented. Minimal storage schemes are introduced to allow for large scale structures to be handled effectively. An accompanying paper demonstrates the performance of the algorithms through solution of several truss and frame problems on an Encore Multimax shared memory multiprocessor computer, for a variable number of processors. The algorithms are particularly suitable for analysis of large structures.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call