Application of a cross decomposition algorithm to a location and allocation problem in distributed systems

  • Authors:
  • Choong Y Lee

  • Affiliations:
  • Gladys A. Kelce School of Business, Pittsburg State University, Pittsburg, Kansas 66762, USA

  • Venue:
  • Computer Communications
  • Year:
  • 1995

Quantified Score

Hi-index 0.24

Visualization

Abstract

In recent years, remarkable progress has been made in the development of computer network systems. The challenging problems that arise in designing a distributed computer network include the determination of sites for the location of computing resources and the allocation of files among these sites. This problem is a complex combinatorial problem that belong to the class of NP-complete problems, where the computation of an optimal solution is still a challenging task. This paper presents a general design model that integrates the choice of computer types into a standard location and allocation problem in distributed computer network systems, and proposes an effective solution algorithm based on cross decomposition. This algorithm unifies Benders decomposition and Lagrangean relaxation into a single framework to reduce the computational difficulty of the problem. Computational results are quite satisfactory and encouraging.