Using a 0-1 integer programming model for automatic static data distribution
This paper describes an automatic data distribution method which deal with both the alignment and the distribution problems in a single optimization phase, as opposed to sequentially solving these two inter-dependent approaches as done by previous work. The core of this work is called the Communicat...
| Autores: | , , |
|---|---|
| Tipo de documento: | artigo |
| Data de publicação: | 1996 |
| País: | España |
| Recursos: | Universitat Politècnica de Catalunya (UPC) |
| Repositório: | UPCommons. Portal del coneixement obert de la UPC |
| Idioma: | inglês |
| OAI Identifier: | oai:upcommons.upc.edu:2117/331039 |
| Acesso em linha: | https://hdl.handle.net/2117/331039 https://dx.doi.org/10.1142/S0129626496000169 |
| Access Level: | Acceso aberto |
| Palavra-chave: | Computational complexity Integer programming Automatic data distribution Automatic parallelization Distributed-memory multiprocessors Computation and data movement costs 0–1 linear programming Complexitat computacional Programació en nombres enters Àrees temàtiques de la UPC::Informàtica::Informàtica teòrica |
| Resumo: | This paper describes an automatic data distribution method which deal with both the alignment and the distribution problems in a single optimization phase, as opposed to sequentially solving these two inter-dependent approaches as done by previous work. The core of this work is called the Communication-Parallelism Graph, which describes the relationships among array dimensions of the same and different array references regarding communication and parallelism. The overall data distribution problem is then formulated as a linear 0–1 integer programming problem, where the objective function to be minimized is the total execution time. The solution is static in the sense that the layout of the arrays does not change during the execution of the program. We also show the feasibility of using this approach to solve the problem in terms of compilation time and quality of the solutions generated. |
|---|