Reinforcement learning to reduce the attack surface in self service cloud computing

Cloud computing offers various services which are analogous to traditional data centers. The on demand supply of resources make this model of utility computing as the platform for many web based services. However security is always a major concern. This thesis proposes a new architecture called Self-service cloud computing with virtual shield (VS) to secure the entire cloud environment. Virtual shield (VS) is designed with the reinforcement learning mechanism to dynamically change the configurations of the client virtual machines (VM) in case of an attack to achieve the required security. This work introduces a novel way to measure the security of the system based on attack surface. The configurations scores generated during the learning process determines the activity of the client. The dynamic configuration of virtual machines in-case of an attack, reduces the attack surface and secures the cloud VM's.