No abstract

No abstract

No abstract

No abstract

Practical solution of the structural analysis problem in a parallel processing environment is investigated through the use of the notion of cheap concurrency and the concept of threads. A thread is a lightweight process or independent instructions executing agent capable of concurrent execution with other threads. Portions of a structural analysis code implemented in C have been parallelized employing the Encore Parallel Threads on an Encore Multimax multiprocessor computer. The issues of racing condition, synchronization, and mapping are considered and discussed. Two synchronization mechanisms, semaphores and monitors, have been employed and compared. Two different mapping strategies have been implemented and studied. Results are reported on the effect of amount and frequency of shared memory access on the speed‐up, the overhead time required for creating threads, and comparison of overall computational time performance using two space truss examples. An overall efficiency of 90–95% was achieved for 11 processors.