Abstract
Tabu search is a simple, yet powerful meta-heuristic based on local search. This paper presents current taxonomy of parallel tabu search algorithms and compares three parallel TS algorithms based on Tabucol, an algorithm for graph coloring. The experiments were performed on two different high performance architectures: an Itanium-based cluster and a shared-memory Sun server. The results are based on graphs available from the DIMACS benchmark suite. © Springer-Verlag Berlin Heidelberg 2007.
Cite
CITATION STYLE
Da̧browski, J. (2007). Parallelization techniques for tabu search. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4699 LNCS, pp. 1126–1135). Springer Verlag. https://doi.org/10.1007/978-3-540-75755-9_130
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
