SLA aware CSP selection and resource monitoring framework for infrastructure as a service cloud
摘要
Over the Internet, cloud computing is becoming increasingly popular as a utility computing model in high-performance computing environments. IoT and recent 5G networks are used to provide these services to users. In the past decade, the market for these services has grown exponentially via the Internet. With this expansion of the cloud, there are now multiple cloud service providers (CSP) offering their services to meet users’ requirements at a lower cost. For the service to meet its requirements, negotiation is required between the CSP and the user. To fill this gap, an intermediary and a policy are required to govern service communication between the user and the CSP. In addition, the intermediary must maintain the quality of service (QoS) delivered to the user by the CSP. This paper proposes a solution to the above problem by using the multi-level broker model to select a cloud service provider by advocating that service level agreements (SLAs) be negotiated, and performance be monitored against service level objectives (SLOs) for infrastructure as a service (IaaS) cloud. The broker model is multi-level, i.e., primary, and secondary brokers. Depending on IaaS SLA parameters, the primary broker employs multi-criteria decision-making (MCDM) and neutrosophic techniques to determine the order of performance by similarity to the ideal solution (NTOPSIS). After getting the ranks of the CSP list, the user selects one CSP and creates a final SLA document between the user and the CSP. Based on this final SLA document, the secondary broker monitors services using the monitoring rules and scenarios defined in the SLO. The performance of the proposed approach has been evaluated by relying on various assessment parameters and compared with state-of-the-art techniques. Based on data obtained from the cloud simulator, the approach has a 10% to 25% lower SLA violation rate than other existing techniques. Resource utilization is between 89 and 85% of customer confidence in cloud services.