A Distributed Differential Evolution Approach for Mapping in a Grid Environment

Abstract

Increase in intensive applications with different computational requirements, coupled with the unification of remote and diverse resources thanks to advances in the widearea network technologies and the low cost of components, have encouraged the development of grid computing. To exploit the promising potentials of geographically distributed resources, effective and efficient mapping algorithms are fundamental. Since the problem of optimally mapping is NP-complete, the development of evolutionary techniques to find near-optimal solutions is welcome. In this paper a distributed system based on Differential Evolution is designed and implemented to face the mapping problem in a grid environment aiming at reducing the degree of use of the grid resources. This system is tested on some different resource allocation scenarios.

Distributed differential evolution with explorative–exploitative population families

“…In Falco et al [17,18,19], in order to solve and image registration problem a distributed DE (here indicated with DDE) has been proposed. This algorithm differs from PDE and IBDDE by the topology it uses.…”

“…In Refs. [17,18,19] a distributed DE has been designed for the image registration problem. In these papers, a computational core acts as a master by collecting the best individuals detected by the various sub-populations running in slave cores.…”

Job Scheduling Algorithms on Grid Computing: State-of- the Art

“…GA and DE are used to schedule jobs on the computational grid [2,20,23]. Evolutionary metaheuristics scheduling mechanisms outperform the grid basic scheduling mechanisms in most cases [50,51].…”

“…Evolutionary metaheuristics scheduling mechanisms outperform the grid basic scheduling mechanisms in most cases [50,51]. De Falco et al in [23] developed a multi objective Differential Evolution mechanism to schedule jobs on computational grid. The objective of the scheduling mechanism is to minimize the resources usage as well as to maximize the grid quality of service requirements.…”

A study on scale factor in distributed differential evolution

“…Following similar logic, Kozlov and Samsonov [23] proposes a distributed DE where the computational cores are arranged according to a ring topology and, during migration, the best individual of a sub-population replaces the oldest member of the neighboring population. In De Falco et al [11][12][13] a distributed DE has been designed for the image registration problem. In these papers, a computational core acts as a master by collecting the best individuals detected by the various subpopulations running in slave cores.…”