PPSA: Profiling and Preventing Security Attacks in Cloud Computing

Cloud computing (CC) is the emerging technology in the world for hosting and delivering services over the internet. It offers a variety of benefits such as cost saving, access to different services without any installation from anywhere and any time by internet, etc. Despite all the advantages offered by CC, this technology still susceptible to security threats. From user point of view, CC seems to be very insecure due to security attacks which threaten it and limit its widespread adoption. More motivations are required to provide more trustworthy solutions to secure the cloud as much as possible and preserve users trust. For that, we propose a new model called "profiling and preventing security attacks", abbreviated to PPSA, to detect and prevent the known attacks as well as the unknown attacks before accessing the cloud services/resources. Unlike the existing solutions, such as intrusion detection systems (IDSs), the proposed solution is able to be deployed in a wide area network in the basis of internet and prevent unknown attacks. In this study, we propose a new solution to profile and prevent security attacks in CC. Then, we define new security factors based on keystroke dynamic. Afterwards, we integrate a machine learning algorithm (classification based on associations) to our proposal in order to profile and predict security attacks and optimize as well the PPSA scheme. Eventually, the proposal is illustrated by a realistic case study.