Predicting Cloud Resource Provisioning using Machine Learning Techniques

Abstract

It can be difficult to provision assets in a cloud environment. Both inadequate and excessive provisioning are bad for the system's overall performance. Researchers are increasingly turning to proactive provisioning techniques that foresee resource needs in advance and set the system up to accommodate such real-time demands. Although proactive provisioning methods are more complicated than reactive provisioning strategies, they offer a generally faster reaction time since decisions about resource provisioning are made in advance of the actual requirement for those resources. The adoption of an analytical framework that anticipates resource requirements is necessary for these proactive provisioning strategies to function well. The provisioning of resources for computational operations in cloud computing is a significant concern. Unsurprisingly, previous research has identified a number of ways to deliver Cloud resources effectively. However, a forecast of the impending computational jobs' future resource consumption is necessary to implement a comprehensive resource provisioning model is required. However, the subject of cloud prediction the use of resources is still in its early stages