Experimental Analysis of DDoS Attacks on OpenStack Cloud Platform

Cloud computing has transformed the IT industry by providing convenient way of accessing resources, services and applications over the Internet. Cloud computing provides infrastructure, platform, software, and other hybrid models as pay-as-you go based services. Customers pay based on usage of particular utility. There are vulnerabilities in the cloud infrastructure which have been easily exploited by the attackers. The well-known Distributed Denial of Service (DDoS) attack is the most eminent attack in this area of computing. DDoS attacks compromise the services of the cloud servers by overwhelming it with huge volume of normal or incomplete traffic making difficult to respond further to the legitimate clients. In cloud computing, virtual machines are created which run through instances. OpenStack is one such cloud platform, which is Open Source and is mostly deployed as Infrastructure as a service (IAAS). This paper presents experimental evaluation of DDoS attack on OpenStack cloud platform. The experimentation was conducted on open source platform of OpenStack as private cloud model. AnonymousDoser, LOIC, and Slowloris were used to launch DDoS attack. These tools flood the cloud with TCP/IP and HTTP packets. Finally, different traffic analysis tools are used to analyze the traffic pattern. The analysis is done on the basis of parameters like resource utilization of CPU, memory and packet drop.