Binomial distribution approach for efficient detection of service level violations in cloud

摘要

Optimum usage of hardware devices, network devices, software resources and consistent quality of service is the aim of cloud computing. Thus, cloud computing is reducing the cost and mounting the revenue for cloud provider. Cloud computing gives the choice and freedom to cloud consumer for the use of dynamic and elastic services as pay-per-use model. In this context, to justify the cost and to verify the quality of services, active Service Level Management is required. Moreover to overcome unrealistic demands from customers and to surpass the occurring violations, a strong and straight forward approach is needed, through which SLA violations can be noticed, detected and distinguished. In this Paper, we have focused on fundamental elements for effective SLA violations detection system. We also propose an approach for binomial distribution for Service level agreement violation detection. This approach gives probability of success and system reliability of cloud service which may be used by service providers to minimize and prevent different violations.