Local Search for the Retrieval Layout Problem

Abstract

An information graph is a data representation for object-oriented databases in which each object is a vertex and each relationship between objects is an edge. The retrieval layout problem is to arrange the storage of an information graph on a physical medium so as to use storage efficiently and to allow rapid navigation along paths in the graph. This paper describes an empirical study of the performance of various local search heuristics for the retrieval layout problem, including local optimization, simulated annealing, tabu search, and genetic algorithms. In addition, the hierarchical hybrid approach is introduced.