The present paper discusses real parallel computations. On the basis of a selected group of dynamic programming algorithms, a number of factors affecting the efficiency of parallel computations such as, e.g., the way of distributing tasks, the interconnection structure between particular elements of the parallel system or the way of organizing of interprocessor communication are analyzed. Computations were implemented in the parallel multitransputer SUPER NODE 1000 system using from 5 to 50 transputers.
2
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
The paper presents an approach to improve the efficiency of some two-level optimization algorithms by their implementation in parallel MIMD multiprocessor systems. Diagonal decomposition dynamic programming and parametric optimization methods are considered, and some concepts of their parallelization are discussed. Results regarding the implementation of computations in a parallel multitransputer system are presented. For the analysed problems, the obtained values of speedup are close to the theoretical maximum values.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.