An Optimized Fuzzy-based Load Balancing in Cloud Computing

In cloud computing, many resources are pooled together to help users operating in a distributed environment collaborate. A load balancer distributes Virtual Machines (VMs) to users in compliance with their required resources and tasks. Existing load balancing algorithms are insufficient for obtaining fast response times and better optimisation of cloud services and their resources when the load increases. Rule-based fuzzy inferences enable optimal resource utilisation by assigning user requests in the most efficient manner. This paper presents an Optimal Fuzzy-based Load Balancing (OFLB) model for efficient resource distribution. The proposed model employs memory, bandwidth, and disc space needs as fuzzy variables and implements categorization-based fuzzy constraints to improve performance. The tasks are assigned to virtual devices based on defined threshold values for membership functions. In the experiments, the OFLB is compared to other extant load-balancing algorithms in terms of memory, bandwidth and disc space utilisation. The analysis of the results shows that the OFLB-based modal improves the efficacy of the cloud system in terms of resource utilization by approximately 18% as compared to existing algorithms that distribute VMs.