论文信息 - Taskgraph Mapping Using a Genetic Algorithm: A Comparison of Fitness Functions

Taskgraph Mapping Using a Genetic Algorithm: A Comparison of Fitness Functions

Abstract A set of parallel processes, represented by an undirected taskgraph with weighted vertices and edges, is mapped to a multicomputer system with an arbitrary interconnection network using a genetic algorithm with parallel populations. Results are presented for mapping thirty taskgraphs onto an nCUBE 2 computer and a comparison is made between several fitness functions on the performance on each mapping.

S. Hurley | S. Hurley

[1] D J Evans,et al. Parallel processing , 1986 .

[2] Heinz Mühlenbein,et al. New solutions to the mapping problem of parallel systems: The evolution approach , 1987, Parallel Comput..

[3] John J. Grefenstette,et al. A Parallel Genetic Algorithm , 1987, ICGA.

[4] John H. Holland,et al. Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

[5] David E. Goldberg,et al. Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[6] Ingo Rechenberg,et al. Evolutionsstrategie : Optimierung technischer Systeme nach Prinzipien der biologischen Evolution , 1973 .

[7] K. Mani Chandy,et al. A comparison of list schedules for parallel processing systems , 1974, Commun. ACM.

[8] Robert Piessens,et al. Load Balancing with Genetic Algorithms , 1992, Parallel Problem Solving from Nature.

[9] J. Ramanujam,et al. Cluster partitioning approaches to mapping parallel programs onto a hypercube , 1987, Parallel Comput..

[10] Jian Xu,et al. Mapping Partitioned Program Modules Onto Multicomputer Nodes Using Simulated Annealing , 1990, ICPP.

[11] Hans-Paul Schwefel,et al. Numerical Optimization of Computer Models , 1982 .

[12] Fred Glover,et al. Tabu Search - Part II , 1989, INFORMS J. Comput..