AND
Download:
|
|
by Paolo Palazzari, Thomas Lippert, Klaus Schilling
http://www.enea.it/hpcn/php/palazzari/Cetraro.ps
Add To MetaCart
Abstract:
Abstract. Systolic processing is a technique [?] to enlarge the range of applications of parallel computers widely, in particular for SIMD and SISAMD (single instruction-- single addressing-- multiple data) architectures. Its hyper-systolic extension can enhance substantially the inter-processor communication speed. The optimization of the actual hyper-systolic communication layout is a non-trivial task if more than 64 processing nodes are involved. Using Simulated Annealing techniques we are able to improve the recently used regular hyper-systolic bases. We find that the speed of interprocessor communication can be increased by about 20 % on the 25 Gflops Quadrics QH4 (with 512 compute nodes) for n 2
Citations
| 9 | Hyper-Systolic Parallel Computing – Lippert, Bode, et al. - 1998 |
| 3 | Data Parallel Algorithms to Solve the Full N-Body Problem – Scholl, J-M - 1993 |
