A parallel adaptive algorithm for the approximate computation of a multi-dimensional integral over an hyperrectangular region is described. This is a more general version of the well known algorithms presented in Genz (1987) and Lapegna and D'Alessio (1993) and it has been developed for an efficient implementation on a MIMD distributed memory multiprocessor. In order to achieve a good scalability, all global communications have been removed from the algorithm. The processor's network has been configured as a multidimensional periodical mesh, to distribute information about the integrand behavior fast enough in order to achieve a good load balancing. Test results on the Intel Touchstone Delta System are given.
Scalability and Load Balancing in Adaptive Algorithms for Multidimensional Integration / M., D'Apuzzo; Lapegna, Marco. - In: PARALLEL COMPUTING. - ISSN 0167-8191. - STAMPA. - 23:8(1997), pp. 1199-1210. [10.1016/S0167-8191(97)00017-3]
Scalability and Load Balancing in Adaptive Algorithms for Multidimensional Integration
LAPEGNA, MARCO
1997
Abstract
A parallel adaptive algorithm for the approximate computation of a multi-dimensional integral over an hyperrectangular region is described. This is a more general version of the well known algorithms presented in Genz (1987) and Lapegna and D'Alessio (1993) and it has been developed for an efficient implementation on a MIMD distributed memory multiprocessor. In order to achieve a good scalability, all global communications have been removed from the algorithm. The processor's network has been configured as a multidimensional periodical mesh, to distribute information about the integrand behavior fast enough in order to achieve a good load balancing. Test results on the Intel Touchstone Delta System are given.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
