An approach to analyse and protect data on Untrusted Cloud Network

The outsourcing of data has gradually increased with the development of cloud computing environment to provide features like pay as you go and auto scaling. This helps small and medium size vendors to expand their business by migrating to cloud. This expansion needs to address important security concern for its customers including data privacy and vulnerability towards the cyber-attacks. One such attack: insider attack continues to worsen the situation of security affecting the confidentiality, integrity and availability of data. This paper reviews different security frameworks designed to detect and prevent abnormalities in the cloud infrastructure. It analyzes two important security hypervisors: TrustVisor and Flicker. Both hypervisors are responsible for providing the information isolation and integrity using piece of logic (PAL) mechanism on the devices where operating system, BIOS and dynamic memory allocation are not recommended reliable. The analysis was done for the execution environment using TPM Remote Attestation, Secure VM launch, Memory Seal Storage based on Storage Root Key (SRK). In the end this paper presents a comparative performance evaluation metrics, for TrustVisor and Flicker hypervisors using eucalyptus cloud software for private cloud. Through this research it was observed that TrustVisor provides high information isolation and integrity as compared to Flicker hypervisor.