Securing Virtual Infrstructure in Cloud Computing using Big Data Analytics

With the popularity of cloud, attacks on cloud are also increasing. For advanced attacks on cloud, attackers choose Virtual Infrastructures to attack as they provide the platform to develop and deploy the applications through Virtual Machines (VMs). Attacks on virtualized infrastructure are increasing day by day. Attackers try to attack on a VM which sometimes leads them to take the unauthorized access of the cloud. There are different techniques to detect and prevent these attacks. Big Data Analytics field is getting better to perform the analysis of attacks efficiently. This paper focuses on Big Data Analytics approaches to detect the advanced Malware attacks in virtualized infrastructure. In this paper we use Big Data Analytics techniques to analyze log files and network traffic to identify anomalies and suspicious activities in virtualized infrastructure in cloud computing. Network logs and application logs are collected periodically from virtual machines (VMs) and stored in the HadoopDistributed File System. We then use Map Reduce Parser and Graph event based correlation to extract the attack features and with the help of logistic regression and belief propagation we will calculate the attacks conditional probabilities and belief in existence of an attack.