A Survey on Resource Scheduling in Cloud Computing Based on Machine Learning and Optimisation Methods

A distributed computing paradigm known as “cloud computing” uses the Internet to deliver computing-that is, processing, storage, services, networks, and applications-in a controlled, virtualized, abstracted, and dynamically demand-driven way. Virtualization, heterogeneity, measured service, pricing, resource pooling, and elasticity are just a few of its standout qualities. The necessity for rigorous resource management has arisen as a result of the cloud computing industry's tremendous growth in demand. That's what resource scheduling addresses. It describes the methodical process of creating a timetable that determines which activities are assigned to which resources. We outline all the significant resource scheduling strategies in this work with the goal of maximizing user Quality of Service (QoS) metrics including cost, makespan, dependability, priority, etc. A thorough investigation of methods that typically enhance user QoS metrics has been carried out. The resource scheduling methods known as deterministic, linear, and evolutionary have been explained, along with a thorough analysis of their differences.