Self-improved algorithm for cloud load balancing under SLA constraints

• Published in: Service oriented computing and applications 2023-12, Vol.17 (4), p.277-291
• Main Authors: Geeta, Koppula, Prasad, V. Kamakshi
• Format: Article
• Language: English
• Subjects: Algorithms; Cloud computing; Computer Appl. in Administrative Data Processing; Computer Science; Computer Systems Organization and Communication Networks; e-Commerce/e-business; IT in Business; Load balancing; Management of Computing and Information Systems; Optimization; Original Research Paper; Rejection rate; Remote computing; Software Engineering/Programming and Operating Systems; Task scheduling; Virtual environments
• ISSN: 1863-2386; 1863-2394
• DOI: 10.1007/s11761-023-00366-8

Because of the remote computing and heterogeneous nature of the infrastructure, load balancing and task scheduling are the most significant difficulties in cloud computing. Many strategies can be used to assign tasks to the appropriate virtual machines (VMs), but balancing the load is the main challenge that arises as a result of changing loads and varying VM characteristics. As a result, resources are used in an imbalanced manner, and system performance suffers. We developed a new Levy flight updated pelican optimization (LFUPO) technique to address these issues. We evaluate load balancing characteristics such as response time, turnaround time, server load, migration cost, and job rejection rate using this method. Based on the SLA constraints, the task rejection rate is also estimated. We are determining the multi-objective function based on these estimated values. Accordingly, the LFUPO approach is used to balance the load with the consideration of tasks ranging in 100, 150, 175, and 200, and VM ranges from 20 to 40. When compared to the prior methodology, this proposed algorithm performs better and distributes the load optimally and effectively.