Abstract
The popularity of the cloud computing has been increased exponentially in recent times. The rapid development of cloud computing technologies has lead to a paradigm shift in the way computing resources are provisioned. Huge number of servers are deployed to serve emerging number of end users everyday. Therefore, the complexity arises to maintain the huge workload as well as to provide expected level of services. Web based applications should have to ensure desired level of quality of services despite dynamic and continuous changes of workload. The main focus of this thesis work is to propose a control system which is implemented with a feedback controlling mechanism known as Proportional Integral Derivative (PID) controller with an existing hybrid controller to dynamically allocate the resources for interactive and non-interactive applications. Traditional approaches are used to take elasticity based decisions such as either monitoring the resource usage or just merely based on Quality of Services (QoS) of solely latency based critical applications specially the webservers. Furthermore, our focus also lies on the batch processing software which is different than the traditional latency critical applications.