Parallel experience on the inverse matrix computation
β Scribed by E. Francomano; A. Pecorella; A. Tortorici Macaluso
- Publisher
- Elsevier Science
- Year
- 1991
- Tongue
- English
- Weight
- 229 KB
- Volume
- 17
- Category
- Article
- ISSN
- 0167-8191
No coin nor oath required. For personal study only.
β¦ Synopsis
In this paper a new structure of computation to evaluate the inverse matrix has been performed, composing in an opportune way the bordering-method and the recursive partitioning method. The scheme reported is based on the block decomposition and on the elimination of the recursion in the process_ The algorithm has been realized in a parallel way on the CRAY Y-MP 8/432 and the multitasking techniques have been used.
π SIMILAR VOLUMES
Consider any known sequential algorithm for matrix multiplication over an arbitrary ring with time complexity O(N a ), where 2 < a [ 3. We show that such an algorithm can be parallelized on a distributed memory parallel computer (DMPC) in O(log N) time by using N a /log N processors. Such a parallel