| Moon, B., Uysal, M., Saltz, J.: Index translation schemes for adaptive computations on distributed memory multicomputers. In: Proceedings of the 9-th International Parallel Processing Symposium, Santa Barbara, California, IEEE Computer Society Press (April 24-28, 1995) 812--819 |
....using a forward Euler explicit time marching scheme in the movement phase of the algorithm. They pointed out that two different levels of data parallelism (i.e. molecules and cells) cause some parallel processing difficulties. Nance et al. 2] parallelized DSMC using the runtime library CHAOS [3] on a 3 D uniform discretized mesh. Robinson and Harvey [4] parallelized DSMC by use of a spatial mesh decomposition over the processors. The domain decomposition is based on a localized load table computed on each processor for its neighbors, which will receive cells donated by the processor if ....
Moon, B., Uysal, M., Saltz, J.: Index translation schemes for adaptive computations on distributed memory multicomputers. In: Proceedings of the 9-th International Parallel Processing Symposium, Santa Barbara, California, IEEE Computer Society Press (April 24-28, 1995) 812--819
.... In order to address these problems, several implementations on concurrent architectures have been presented [11, 17, 18, 20, 23, 32] Parallel scalability analysis has been presented in [38] Load balancing and related performance optimizations for particle based simulations are discussed in [17, 22, 30, 35, 36, 39]. The use of automatic granularity control for load balancing of DSMC simulations is presented in [28] The scalability of similar techniques has been studied in [10, 12] This paper considers the parallel scalability of the DSMC method, from an applicationand implementation independent ....
B. Moon, M. Uysal and J. Saltz, " Index Translation Schemes for Adaptive Computations on Distributed Memory Multicomputers," University of Maryland Technical Report CS-TR-3428, 1995.
....application, the table is replicated on each processor because the problem size is relatively small. However, memory considerations make it clear that it is not always feasible to replicate the table, so the translation table must be distributed across processors in some applications. Moon et al. [20] have recently developed new index translation schemes which use software caching techniques so that extra memory can be exploited adaptively for changeable data access patterns and communication latency can be avoided. However, as mentioned above, simple table replication was utilized for this ....
Moon, B., Uysal, M., and Saltz, J., "Index Translation Schemes for Adaptive Computations on Distributed Memory Multicomputers," Proceedings of the Ninth International Parallel Processing Symposium, IEEE Computer Society Press, Santa Barbara, CA, April 1995 (to be published).
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC