Security solutions in Cloud through customized IDS configuration at VM level

Abstract

Security is one of the top concerns in the adoption of Cloud Computing. Due to the distributed architecture, assessed through the internet cloud architecture are exposed to critical network attacks. Intrusion Detection System is used to ensure network security and early detection of possible attacks. When deployed in Cloud architecture, Intrusion Detection System has to confront issues like high amounts of data to be processed due to high network traffic and high false alarm rates due to dynamic nature of cloud. This paper will provide overview of working of IDS and its deployment in cloud. We propose a novel algorithm that works in accordance with the Virtual Machine’s feature and network behavior to select signatures for configuring intrusion detection system at virtual machine level. This system aims at providing high detection accuracy with low false positive rate at affordable computational cost. We will use snort, a signature based detection tool to detect the known attacks. IDS at each instance will be configured with selected signatures in the cloud environment. The system developed is evaluated for IDS accuracy and resources required.