A Novel Cloud Intrusion Detection System Using Feature Selection and Classification

1. Introduction

Infrastructure as a Service (IaaS) is a cloud service provisioning model aimed at providing the users access to cloud resources, such as storage and computing freely from various digital devices through the Internet. The benefits of deploying applications in the cloud are manifold including better economics through use of shared computing resources, much lower upfront infrastructure costs, and on-demand elastic provisioning of computing nodes to fit transient requirements (Kholidy & Baiardi, 2012).

Thus, for the applications that have a high degree of variable demand for resources, the cloud based infrastructure model offers an efficient and cost-effective method to provide the resources used, while minimizing the overall cost. Virtualization within data-centers had been a key enabler for dynamic provisioning model of computing resources in cloud. However, virtualization benefits are limited by the security challenges it brings along and newer types of attacks. Recently, service oriented attackers are increasing in the Internet. Different kind of attacks such as Denial of Service (DoS) attacks and Distributed Denial of Service (DDoS) attacks are growing in the Internet. These attacks reduce the system performance drastically. Since Cloud computing is to a large extent dependent on internet, a good security mechanism is necessary to support the cloud applications. Recently, Cloud Intrusion Detection System (CIDS) has been proposed by many researchers for improving the cloud services. CIDS identifies anomaly behavior by monitoring and analyzing the user traffic. It provides security to distributed cloud applications from the attackers (Oktay & Sahingoz, 2013; Anand, 2012). Such a system can be developed by using effective feature selection and classification algorithms which can primarily identify the attackers and normal users accurately. In addition, the attack types and the attack patterns can also be identified using proper use of feature selection techniques with intelligent classification algorithms.