A Unified Multi-Objective Adaptive Task Scheduling for VM Optimization in Heterogeneous Cloud Data Center

In light of the swift progress of cloud technology, substantial cloud computing centers have surfaced to cater to the escalating demand in the business realm. These centers are designed to cater to the increasing demands of data processing and machine learning, with their vast number of computing resources, efficient resource and energy management becomes crucial in order to reduce costs and minimize environmental impact. Energy efficiency in cloud computing is a critical concern as the demand for cloud services continues to grow. Energy efficiency in cloud computing is not only a technological imperative but also a pivotal step towards a more sustainable and environmentally responsible future. With the increasing environmental awareness and the need to reduce operational costs, optimizing energy consumption in data centres and cloud infrastructure is essential. The Virtual Machine Consolidation (VMMC) is a strategy to optimize power utilization in the Data Center of Cloud by optimizing hardware and software, consolidating virtual machines, scheduling workloads, and integrating renewable energy. In this paper, we proposed the ARMS (Adaptive Resource Management Systems) framework to a mixture of the Autoregressive Integrated Moving Average (ARIMA) model with Random Forest Algorithm (RFA). The experiment results indicate that the benchmark methods are outperformed in terms of Energy Efficiency, Cost, Makespan, SLAV, and Number of Iteration etc.