Service-level agreement aware energy-efficient load balancing in cloud using hybrid optimization model

• Published in: Service oriented computing and applications Vol. 17; no. 2; pp. 77 - 91
• Main Authors: Kodli, Shilpa B., Terdal, Sujata
• Format: Journal Article
• Language: English
• Published: London Springer London 2023; Springer Nature B.V
• Subjects: Cloud computing; Computer Appl. in Administrative Data Processing; Computer centers; Computer Science; Computer Systems Organization and Communication Networks; Correlation coefficients; Data centers; e-Commerce/e-business; Energy consumption; Heterogeneity; IT in Business; Load balancing; Management of Computing and Information Systems; Optimization; Optimization models; Original Research Paper; Resource allocation; Resource management; Software Engineering/Programming and Operating Systems; Virtual environments; Workload; Workloads; Central manager; VM allotment; Cloud computing; SMI-CSO model; Resource allocation
• ISSN: 1863-2386; 1863-2394
• DOI: 10.1007/s11761-023-00359-7

Cloud computing (CC) is the most promising area of research for the previous few years, which offers shared resource power on demand. Virtualization is a major characteristic of CC, which entails the formation of several virtual machines (VMs) on a particular physical computer. The VM migration mechanism is aided by varied cloud service providers (CSPs) for managing resources. Nevertheless, if the distribution of resources is not handled accurately the VM allotment cannot be optimal. In addition, due to the enhanced number of data centers, energy consumption is high. Besides, resource management has become a major issue, due to the heterogeneity of resources. Due to the lack of resources on physical computers, several VM migrations cause data centers to perform worse (PM). Consequently, it is essential to drastically cut on energy use and the amount of VM migrations without violating SLA. Hence, this work goes ahead with three modules namely, (a) workload submission, (b) central manager (CM), and (c) migration. During workload submission, the tasks are submitted by the cloud users. In the second phase, resource allocation and migration take place. During migration, a new safety constraint is introduced using the improved Pearson correlation coefficient (IPCC) model. Here, hybrid optimization model named spider monkey-induced cat swarm optimization (SMI-CSO) is incorporated to select the VM and to allocate the resources. Further, enhanced correlation-based VM placement and global agent (GA) apply load balancing. Finally, VM performs actual migration from the schedule received from GA. Finally, the advantage of the suggested scheme is proven by wide-ranging metrics.