Trust Analysis For Defending Flooding Attacks In SDN Based Cloud

Abstract

Distributed Denial of Service (DDoS) attack is the explicit attempt to exhaust the victim’s bandwidth thereby disrupting user’s access to services. Due to growing technologies, in the cloud computing environment, SDN-based cloud has generated the new trails to defeat the DDoS attack. The proposed objective focuses on communication protocol based attack, i.e. TCP SYN Flood attack. The chosen attack is executed in the proposed distributed environment which comprises of a number of hosts connected via multiple OpenFlow controllers. This attack is structured in such a way so as to mimic the actual ’on-off’ property of real world attackers. The proposed approach involves trust based evaluation, in contrast to the intrusion detection done using secured trust evaluation metrics. A trust algorithm is implemented as a result to fight the malicious nodes and hence prevent secure private data to travel through such hosts. However in previous work, the outermost plane,i.e., data plane has been utilized. In this work, the data plane is separated from control plane thereby working with the latter one providing a centralized point of control in the network.