Title :
Optimization of large structures on multiprocessor machines
Author :
Adeli, Hojjat ; Kamal, Osama
Author_Institution :
Dept. of Civil Eng., Ohio State Univ., Columbus, OH, USA
fDate :
30 Sep-2 Oct 1990
Abstract :
For the minimum-weight design of structures, the authors present parallel algorithms that are particularly suitable for optimization of large structures such as space stations. The parallel algorithms have been implemented in C on an Encore Multimax shared-memory multiprocessor computer using the concept of cheap concurrency and threads (H. Adeli and O. Kamal, 1989). Emphasis is directed toward parallelizing each computational step of the solution process. Parallel algorithms are developed to ensure the best concurrent performance and speedup within each step. A partitioning algorithm is used to achieve the best workload balance during the solution of the generalized displacements
Keywords :
concurrency control; multiprocessing systems; parallel algorithms; C; Encore Multimax shared-memory multiprocessor computer; concurrency; generalized displacements; large structures optimisation; minimum-weight design; multiprocessor machines; parallel algorithms; partitioning algorithm; space stations; workload balance; Algorithm design and analysis; Assembly; Boundary conditions; Equations; Flowcharts; Iterative algorithms; Performance analysis; Stress; Vectors; Yarn;
Conference_Titel :
Distributed Computing Systems, 1990. Proceedings., Second IEEE Workshop on Future Trends of
Conference_Location :
Cairo
Print_ISBN :
0-8186-2088-9
DOI :
10.1109/FTDCS.1990.138335