Hybrid optimal load balancing with modified fuzzy clustering-based virtual machine categorization

In cloud computing, load balancing is crucial for effective resource management. Keeping servers from getting overworked entails dividing up incoming network traffic or computational tasks among several servers. Better resource management, increased throughput, and quicker reaction times result from this. Several heuristic and metaheuristic techniques have been used to disperse the load across the available virtual machines. Researchers have worked very hard to find a solution for the load balancing issue. This paper uses the following stages to develop a novel load-balancing model with optimization assistance: Virtual Machine (VM) classification, load balancing, and replica management are the three main processes. For the VM classification process, a modified version of the fuzzy clustering approach is suggested. For the load balancing procedure, the COOT Insisted Bald Eagle Search (COOTIBES) model is suggested. This optimization-assisted load balancing takes into account a number of limitations, including frequency, makespan, memory usage, resource utilization, and execution time. Additionally, the suggested COOTIBES algorithm manages replicas while taking load, put cost, and storage cost into account. Lastly, using various performance indicators, the suggested work’s performance is contrasted with traditional models. While the Inquisitive Genetic Algorithm with Grey Wolf Optimization Algorithm (IG-GWO), Ant Colony Optimization for Continuous Domains (ACOR), Life Choice-Based Optimizer (LCBO), Cat Swarm Optimization (CSO), Genetic Algorithm Combined with First Come First Serve + Genetic Algorithm Combined with Round Robin (GA-FCFS[Formula: see text][Formula: see text][Formula: see text]GA-RR), Jellyfish Optimization (JFO), and Namib Beetle Optimization (NBO) offer bigger makespan values, the COOTIBES scheme gives a smaller makespan of 180 for a task count of 2000.