Resumo

DIAZ LOPEZ, Ernesto; MARTINEZ PRIETO, Adrian  e  GALVEZ LIO, Daniel. An implementation for the meta-heuristic “Variable Mesh Optimization” on CUDA architecture. Rev cuba cienc informat [online]. 2016, vol.10, n.3, pp. 42-56. ISSN 2227-1899.

ABSTRACT In the field of optimization, both academic and industrial problems that occur daily are often complicated and NP-hard. The computational solution thereof can’t be performed in an accurate manner within a reasonable time, and use of computer hardware resources for this solution increases significantly. To address this issue, the design of the solution should be based on the joint use of advanced combinatorial optimization approaches and methods of large-scale parallelismIn recent decades, in the field of artificial intelligence, population based meta-heuristics represent a powerful tool for solving complex problems. The objective of this work is to expose the results obtained from the application of parallel techniques using the CUDA architecture to the meta-heuristic "Variable Mesh Optimization", with the purpose of achieving an improvement in the temporary cost. The most expensive steps of the algorithm in the process of expansion and contraction of the mesh to obtain a new initial population are implemented in parallel. The results were validated using a sequential version of the meta-heuristics. The runtimes of both algorithms are compared, showing substantial improvements in the parallel version using CUDA.

Palavras-chave : Meta-heuristic; Variable Mesh Optimization; Parallel Programming; CUDA.

        · resumo em Espanhol     · texto em Espanhol     · Espanhol ( pdf )