Magn'us M. Halld'orsson y
|
Abstract:
The general block distribution of a matrix is a rectilinear partition of the matrix into orthogonal blocks such that the maximum sum of the elements within a single block is minimized. This corresponds to partitioning the matrix onto parallel processors so as to minimize processor load while maintaining regular communication patterns. Applications of the problem include various parallel sparse matrix computations, compilers for high-performance languages, particle in cell computations, video and image compression, and simulations associated with a communication network. We analyze the performance guarantee of a natural and practical heuristic based on iterative refinement, which has previously been shown to give good empirical results. When p
Citations
| 57 | Extending HPF for Advanced Data Parallel Applications – Chapman, Mehrotra, et al. - 1994 |
| 35 | Rectilinear partitioning of irregular data parallel computations – Nicol - 1991 |
| 25 | Efficient partitioning of sequences – Olstad, Manne - 1995 |
| 20 | Efficient array partitioning – Khanna, Muthukrishnan, et al. - 1997 |
| 14 | Load Balancing in Parallel Sparse Matrix Computations – Manne - 1993 |
| 10 | Partitioning an array onto a mesh of processors – Manne, Sorevik - 1996 |
| 9 | On the complexity of the generalized block distribution – Grigni, Manne - 1996 |
| 8 | Optimal algorithms for partitioning trees and locating p-centers in trees – Frederickson - 1990 |
| 4 | Partitioning a matrix to minimize the maximum cost – Mingozzi, Ricciardelli, et al. - 1995 |
| 2 | Performance Fortran Language Specification 2.0 – High - 1997 |
| 2 | On discrete spacings and the Bose-Einstein distribution, in Contributions to Probability and Statistics in the Honor of Gunnar Blom – Holst - 1985 |
| 2 | acs, Solution to Problem 6386 – Tak' - 1983 |
