Title: Resource management policies to increase provider's gain in a cloud coalition

Authors: Nassima Bouchareb; Nacer Eddine Zarour; Samir Aknine

Addresses: Faculty of New Technologies of Information and Communication, University of Abdelhamid Mehri, Constantine 2, Nouvelle Ville Ali Mendjeli Box 67A, Constantine, Algeria ' Faculty of New Technologies of Information and Communication, University of Abdelhamid Mehri, Constantine 2, Nouvelle Ville Ali Mendjeli Box 67A, Constantine, Algeria ' Department of Computer Science, University Claude Bernard, Lyon 1, 43, Bd of 11 November 1918, 69622 Villeurbanne Cedex, Lyon, France

Abstract: Cloud computing is a new trend after grids. The resource allocation in such dynamic environments presents numerous challenges. Unfortunately, the maximisation of accepted user's requests and at the same time the reduction of provider's costs are a conflicting problem. In this paper, we propose an agent-based mechanism to automatically manage cloud resources in order to simultaneously achieve the maximum of accepted requests with minimum cost and to reduce as much as possible the amount of energy consumption. We present new strategies that help in the decision-making process. These strategies increase provider's profit (forming or leaving coalitions, accepting or rejecting an offer and calculating the cloud trust). Finally, this paper presents a case study to illustrate our resource management mechanism in cloud computing. Also, simulation results indicate that the proposed policies enhance the provider's profit, and give a smaller number of rejected virtual machines requests in the coalition environment.

Keywords: cloud computing; resource management; green computing; multi-agent systems; MAS; agent-based systems; cloud coalition; cloud resources; cloud trust; simulation.

DOI: 10.1504/IJGUC.2016.080184

International Journal of Grid and Utility Computing, 2016 Vol.7 No.3, pp.163 - 176

Received: 14 Apr 2014
Accepted: 03 Nov 2014
Published online: 07 Nov 2016 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article