Energy-Aware Dynamic Virtual Machine Scheduling in Cloud Computing: A Survey
Kashav Ajmera, Tribhuwan Kumar Tewari, Vivek Kumar Singh, Vikash, Pawan Kumar Upadhyay · 2023
Dynamic virtual machine scheduling is a crucial research area in cloud computing, to optimize resource utilization, reduce energy consumption, and ensuring performance and availability by dynamically managing and scheduling virtual machines (VMs) across servers. Effective dynamic VM scheduling can significantly impact the efficiency and sustainability of data centers by efficiently allocating resources based on workload demand. However, the dynamic nature of workload demand presents one of the primary challenges in dynamic VM scheduling. Sudden spikes or drops in server resource usage can lead to severe SLA violation concerns and energy inefficiency. To mitigate these challenges, scheduling algorithms need to balance the workload demand across servers while ensuring that servers are not overloaded or underloaded. In this work, we present a comprehensive comparison of state-of-the-art techniques for dynamic VM scheduling, evaluating their performance based on several key factors, including objective function, approach, the technique of VM scheduling, VM placement approach, VM selection approach, and server underload and overload detection techniques. Our comparison highlights the strengths and weaknesses of different approaches and provides insights into the design of efficient dynamic VM scheduling algorithms. The findings of this work can guide the development of novel techniques that address the challenges of dynamic VM scheduling and improve the efficiency and sustainability of data centers.