Comparative analysis of consumer profile-based methods to predict SLA violation

A Service Level Agreement (SLA) is a contract between a service provider and a consumer which specifies in detail the level of service expected from the service provider, obligations, commitment and objectives. In the cloud computing environment, both the cloud provider and the cloud consumer want to know of a likely service violation before the actual violation occurs and to adjust the scaling of the cloud resources appropriately. A consumer´s previous resource usage profile is a key element in determining the possibility of service violation in the cloud computing environment, which has not been an area of research focus so far. In this paper, we analyze and compare QoS prediction by considering the consumer´s previous resource usage profile in various conditions. From comparative analysis, we observe that by combining a consumer´s previous resource usage profile history along with the previous resource usage profile history of its nearest neighbors, we obtain an optimal result.