SLA Based an Efficient and Reliable Resource Provisioning in Private Cloud

Abstract

Cloud computing has turned into the most prevalent distributed computing environment in light of the fact that it doesn't obliges administration and controlling on more level execution at client level. Nonetheless, effective resource provisioning is a key test for cloud computing and determining such sort of issue can diminish under or over utilization of resources, increment client fulfillment by serving more clients during peak hours, decrease execution cost for suppliers and service cost for clients.

Existing works on cloud computing focuses estimation the accurate capacity needs, static or dynamic VM (Virtual Machine) creation and scheduling. But significant amount of time is required to create and destroy VMs which could be used to serve more user requests. In this thesis, an QoS (Quality of Service) aware VM provisioning mechanism is developed that ensures efficient utilization of the system resources. The VM for similar type of requests has been recycled so that the VM creation time can be minimized and used to serve more user requests. In the proposed model, QoS is ensured by serving all the tasks within the requirements described in SLA.

The advances of virtualization technology in Data Centres lead to the development of features such as High Availability and Fault Tolerance. This thesis also presents the enablement solution that covers VM migration in private clouds.