RETRACTED ARTICLE: Modified adaptive neuro fuzzy inference system based load balancing for virtual machine with security in cloud computing environment

Abstract

In a heterogeneous environment, computation over internet is provided by a popular paradigm called cloud computing. In a cloud heterogeneous environment, service providing has various difficulties. Based on service type, difficulties differ. On cloud server, high load is produced by huge amount of request from various users for accessing various applications. Security and balancing of load are major concerns. In cloud environment, NP-hard optimization problem corresponds to load balancing. Dynamic load balancing is handled by various methods. They are designed for enhancing workload distribution process between nodes. Overload avoidance, minimization of average response time, data processing time and optimum utilization of resources are the major aim of those methods. An optimal load balancing technique should improve the turnaround time and maximum CPU utilization. Because of its opaqueness nature of cloud, security is a biggest challenge. According to Forbes, with introduction of General Data Protection Regulation security in cloud continue to be an issue with cloud computing. In the existing system, a fuzzy based hybrid load balancing algorithm is utilized and the results provided are not satisfactory. There are opportunities for improving CPU utilization and turnaround time and in terms of security. In this proposed research work, dynamic load balancing in a heterogeneous environment is handled by Modified Adaptive Neuro Fuzzy Inference System (MANFIS). Parameters of MANFIS are optimized by introducing Fire-fly Algorithm. Security is imposed on user authentication by using the Enhanced Elliptic Curve Cryptography. This is a password-less mechanism to authenticate users. The proposed work attains satisfactory results by proper resource utilization. An experimental result shows that proposed work exhibits better performance by improving the turnaround time and maximizing the CPU utilization and providing secured access to data.