Title: A hybrid load balancing algorithm for P2P-cloud system aware of constraints optimisation of cost and reliability criteria
Authors: Sofiane Mounine Hemam; Ouassila Hioual
Addresses: Department of Mathematics and Computer Science, Abbes Laghrour University, Khenchela – 40004, Algeria; ICOSI Laboratory, Khenchela – 40004, Algeria ' Department of Mathematics and Computer Science, Abbes Laghrour University, Khenchela – 40004, Algeria; LIRE Laboratory, Contantine – 25000, Algeria
Abstract: P2P-cloud, also known as volunteer cloud, is a new paradigm which aims at moving volunteer computing towards the cloud. It could be built from reliable resources found in the cloud with a cost, and from unreliable resources found in many private homes with essentially a very low cost. However, the reliability and cost are contradictory related and have a negative impact on load between resources. This paper proposes a hybrid load balancing algorithm, which takes into consideration user requirements (cost and reliability) and system performance (the load) in the context of volunteer cloud. The proposed algorithm regulates the load between load balancers in a distributed manner, and between resources in a centralised manner by optimising cost and reliability criteria. Our approach is evaluated through simulation by using PeerSim simulator. The experimental results show that we can effectively achieve good performance (load balancing between resources) and improve the user requirements.
Keywords: cloud system; P2P system; dynamicity; volunteer cloud; reliability; failure probability; load balancing; Pareto optimal; TOPSIS; multi-objective optimisation; cost.
International Journal of Internet Protocol Technology, 2017 Vol.10 No.2, pp.99 - 114
Received: 17 Dec 2015
Accepted: 08 Dec 2016
Published online: 27 Jun 2017 *