Authors: S. Bharath Bhushan; Pradeep Reddy
Addresses: School of Information Technology and Engineering, VIT University, Vellore, India ' School of Information Technology and Engineering, VIT University, Vellore, India
Abstract: Cloud computing is an evolving standard that delivers computing and storage possessions as a service over the internet. One of the major key concerns is load balancing in multi cloud domain, which is NP-hard optimisation problem. In this paper, we proposed a biogeography-based load balancing algorithm (BB-LBA), which balances the load across geographically distributed datacentres and virtual machines for maximising the throughput. We derive Markov models for biogeography-based load balancing in multi cloud domain with migration and mutation operators. A comparative experiments with GA, PSO and HBB-LB is conducted, and the results show that our proposed BB-LBA algorithm outperforms the other in term of response time, number of task migrations and waiting time of tasks in queue.
Keywords: biogeography-based optimisation; BBO; cloud computing; evolutionary algorithms; EAs; load balancing; Markov models; distributed data centres; virtual machines; throughput; multi cloud domain; response time; task migration; waiting time.
International Journal of Internet Protocol Technology, 2016 Vol.9 No.2/3, pp.100 - 106
Received: 19 Jan 2016
Accepted: 26 Mar 2016
Published online: 30 Sep 2016 *