Authors: Jorge M. Cortés-Mendoza; Andrei Tchernykh; Ana-Maria Simionovici; Pascal Bouvry; Sergio Nesmachnow; Bernabé Dorronsoro; Loic Didelot
Addresses: Computer Science Department, CICESE Research Center, Ensenada, Baja California, México ' Computer Science Department, CICESE Research Center, Ensenada, Baja California, México ' Computer Science and Communications Research Unit, University of Luxembourg, Luxembourg ' Computer Science and Communications Research Unit, University of Luxembourg, Luxembourg ' Faculty of Engineering, Universidad de la República, Montevideo, Uruguay ' Department of Computer Engineering, Universidad de Cádiz, Province of Cádiz, Andalusia, Spain ' MIXvoip S.a., Sandweiler, Luxembourg
Abstract: Voice over IP (VoIP) is very fast growing technology for the delivery of voice communications and multimedia data over internet with lower cost. Early technical solutions mirrored the architecture of the legacy telephone network. Now, they have adopted the concept of distributed cloud VoIP. These solutions typically allow dynamic interconnection between users on any domains. However, providers face challenges to use infrastructure in the best efficient and cost-effective ways. Hence, efficient scheduling and load balancing algorithms are a fundamental part of this approach, especially in presence of the uncertainty of a very dynamic and unpredictable environment. In this paper, we formulate the problem of dynamic scheduling of VoIP services in distributed cloud environments and propose a model for bi-objective optimisation. We consider it as the special case of the bin packing problem, and discuss solutions for provider cost optimisation while ensuring quality of service.
Keywords: cloud computing; load balancing; provider cost minimisation; QoS; quality of service; VoIP; voice over IP; service models; multi-objective scheduling; cloud infrastructure; bin packing problem; provider cost optimisation.
International Journal of Metaheuristics, 2015 Vol.4 No.2, pp.185 - 203
Available online: 18 Jan 2016 *Full-text access for editors Access for subscribers Purchase this article Comment on this article