A Survey of Methods and Architectures for A Proactive Auto-scaling and Energy-Efficient VM Allocation Framework Using an Online Multi-Resource Capsule Shuffle Attention Network for Cloud Data Centres

Abstract

Cloud computing has become a vital component of modern IT infrastructure, providing scalable resources, distributed storage, and flexible service deployment through cloud data centres. These centres support a wide range of applications, from enterprise systems to advanced artificial intelligence platforms, requiring efficient resource management to maintain performance and reliability. However, the rapid growth of cloud services has introduced challenges such as dynamic workload handling, resource allocation, and high energy consumption. One major concern is the efficient allocation of virtual machines (VMs) to physical servers, as poor allocation can lead to underutilization, increased power usage, higher operational costs, and environmental impacts due to carbon emissions. As a result, energy-efficient resource management has become a key research focus for sustainable cloud operations. Auto-scaling mechanisms play an important role by dynamically adjusting resources based on workload demands. While traditional reactive approaches respond after performance degradation, proactive auto-scaling uses predictive models to anticipate workload changes and allocate resources in advance. This approach improves system responsiveness, reduces SLA violations, and enhances overall efficiency in cloud data centres.