Authors: Rahul Bhalerao; Madhur Bajpai; Avinash Jadhao; Shashikant Zalte; Jibi Abraham
Addresses: College of Engineering, Pune, Maharashtra, 411005, India ' College of Engineering, Pune, Maharashtra, 411005, India ' College of Engineering, Pune, Maharashtra, 411005, India ' College of Engineering, Pune, Maharashtra, 411005, India ' College of Engineering, Pune, Maharashtra, 411005, India
Abstract: Cloud computing is the latest trend in the world of computers. Virtualisation is the building block for cloud. In IAAS, cloud resources are allocated to the user in the form of virtual machines (VMs). VMs generally have lower resources and have a good chance of getting overloaded. Overload can happen in terms of CPU, memory or IO. All these three resources are divided among the different VMs running on a physical machine. So even if a physical machine is not overloaded, a VM running on it may get overloaded. Considering the nature of the loads, this paper proposes a load balancing solution on basis of CPU, memory and IO using process migration between the VMs of the cloud. The approach balances the load in such a way that the use of energy efficiency techniques like DVFS can be maximised. Due to the multi-tenant model of the cloud, a lot of new security issues have come up. Process migration also presents new security issues. This paper identifies such security issues and proposes solution to them. A proof of concept is provided which demonstrates the feasibility of the approach by proving that improved response times can be achieved even at lower energy consumptions.
Keywords: cloud computing; virtualisation; load balancing; cloud security; energy efficiency; DVFS; process migration; checkpoint restart approach; predictive analysis; distributed computing; data centre hot spots; virtual machines; response times; energy consumption.
International Journal of Cloud Computing, 2014 Vol.3 No.3, pp.288 - 314
Available online: 14 Sep 2014 *Full-text access for editors Access for subscribers Purchase this article Comment on this article