Detecting and Preventing of Attacks in Cloud Computing Using Hybrid Algorithm

Abstract

Cloud computing is the technology that is currently used to provide users with infrastructure, platform, and software services effectively. Under this system, Platform as a Service (PaaS) offers a medium headed for a web develop-ment platform that uniformly distributes the requests and resources. Hackers using Denial of service (DoS) and Distributed Denial of Service (DDoS) attacks abruptly interrupt these requests. Even though several existing methods like sig-nature-based, statistical anomaly-based, and stateful protocol analysis are avail-able, they are not sufficient enough to get rid of Denial of service (DoS) and Distributed Denial of Service (DDoS) attacks and hence there is a great need for a definite algorithm. Concerning this issue, we propose an improved hybrid algorithm which is a combination of Multivariate correlation analysis, Spearman coefficient, and mitigation technique. It can easily differentiate common traffic and attack traffic. Not only that, it greatly helps the network to distribute the resources only for authenticated requests. The effects of comparing with the nor-malized information have shown an extra encouraging detection accuracy of 99% for the numerous DoS attack as well as DDoS attacks.