Integrating the EGC, EF, and ECS Trio Approaches to Ensure Security and Load Balancing in the Cloud

Abstract

According to data protection studies, "Distributed Denial-of-Service (DDoS)" threats have cost governments and businesses throughout the globe a large number of financial resources. Despite this, the existing practices fall short of the standards set by "Cloud Computing (CC)" monitoring technology. They ignore the "Intrusion Detection Systems (IDS)" techniques, which take advantage of the CC's multiple tenants and elasticity qualities, and also the hardware limitations. Attackers are finding increasing ways to effectively exploit them because of their rising complexity. DDoS assaults of this scale have never been observed online before 2018. As online services get more popular, so does the amount of DDoS assaults and malevolent hackers leading to terrible. Numerous IDS for DDoS are already in place to address this problem. One of the most challenging aspects of virtualization is establishing a "Trust Model (TM)" between the many "Virtual Machines (VMs)". The lack of a standard formulation for generating a TM would be the primary reason. As a consequence, the integrity of every VM might not have been recognized by an independent trust, which might lead to a decrease in trust value. In this research for TM creation, "Enhanced Graph Based Clustering (EGC)" is proposed, while "Enhanced Fuzzy (EF)" is used for detecting attacks, and the "Enhanced Cuckoo Search (ECS)" method is used to find the ideal "Load Balancing (LB)" distribution. By creating a new TM, the proposed (EGC-EF-ECS) system strengthens trust value. To expand the CC model's stability, it optimizes attacker recognition percentage and makes better use of resources by restricting each VM's processing, bandwidth, and storage requirements. The proposed EGC-EF-ECS outperformed the previously used BPA-SAB, and DCRI-RI approaches in terms of the "Intrusion-Detection-Rate (IDR)", "Load-Balancing-Efficiency (LBE)", and "Data-Accessing-Time (DAT)" evaluation metrics.