Enhanced two stage heuristics algorithm for VM scheduling
S. Selvarani, Anitha Julian, E. Iniya Nehru · 2016
In cloud data center deployment of Virtual Machine (VM) reduces the active number of Physical Servers. An efficient VM Scheduling scheme requires to reduce the energy consumption as well as to improve the network performance. In contrast to the existing solution, VM placement and VM migration is proposed in this work. In VM placement Physical Machine (PM) resources are formulated as Bin Packing Problem (BPP), and network traffic as Quadratic Assignment Problem (QAP). Both the BPP and QAP are NP hard problem, so the greedy approach is taken to solve the problem. After the successful VM placement the VM migration scheme analyzes the Maximum Link Utilization (MLU), Round Trip Time (RTT) and Loss Rate (RT). The design of CloudSim simulator to implement the greedy approach is presented and simulations are run over data center resources and network elements. The simulation results when compared with existing algorithms show remarkable improvements.