SMLHADC: Security Model for Load Harmonization and Anomaly Detection in Cloud

Abstract

Cloud computing (CC) emerged as one of the important utility such as water and electricity bills, where the user has to pay for whatever quantity they have utilized for their tasks. As cloud users are increasing; so the crucial challenges for the cloud service provider (CSP) is to balance the load among the resources shared by the end-users and curb increased security risks, misuse, or malicious attacks in cloud computing. The functionality of load balancing is divided into two functions, first, there will be allocation of resources and the second is the provisioning of resources along with task scheduling in the distributed system such as CC. Although many state-of-the-art approaches are available in the literature which provides load balancing and better resource utilization. In this paper, we build a dynamic prediction approach for cloud resource usage, discuss a load balancing technique, and identify sudden spikes and failures that are causes of the anomaly using a reactive and proactive approach. The final result demonstrates a comparative analysis of deep learning models for the prediction of cloud resource utilization, insight into dynamic load balancing technique and anomaly detection.