Abstract
Conference Title: 2016 18th International Conference on Advanced Communication Technology (ICACT) Conference Start Date: 2016, Jan. 31 Conference End Date: 2016, Feb. 3 Conference Location: Pyeongchang Kwangwoon Do, South Korea This paper presents a cost-effective, scalable and energy-efficient cloud resource allocation model for real-time seamless access and processing of monitored body sensor data. In particular, two methods were used to solve the VM resource allocation problem The first method uses a linear programming (LP) model of quantitatively optimizing VM allocation using cost objectives and constraints on the resource utilization condition, CPU utilization, energy consumption and delay of services. The second approach solves the VM resource allocation problem using different heuristics. We studied existing heuristics and modify them to fit our situation. We tested the workload of body sensor data on Amazon Cloud EC2. Then we create a large simulation scenario and compare our proposed method with other approaches.