Uma estratégia para assegurar a confidencialidade de dados armazenados em nuvem

Abstract

Large amounts of confidential data stored on servers in the cloud is a trend for companies looking for opportunities to reduce costs and increase the availability of their digital services. However, in cloud computing environments data control is no longer belongs to the data owner and the control belongs to the service provider, which provides new challenges related to privacy, security, and confidentiality. In this context, privacy and security solutions for data stored in the cloud using encryption, data fragmentation or a combination of both have been proposed as best existing techniques in the scientific literature. Despite this, problems related to the effectiveness of these techniques in relation to the attacks, loss or theft of data have occurred in recent years, causing millions of dollars of damages to companies and clients. In this thesis, we present a new approach, called QSM-EXTRACTION, to ensure the confidentiality of data in cloud storage services. The science behind this approach uses concepts from Hegel’s Doctrine of Being. The QSM-EXTRACTION strategy is based on the fragmentation of a digital file into fragments called information objects, on the decomposition of these objects through the extraction of their characteristics (Quality, Quantity and Measure) and the dispersion of these characteristics in different storage services in Cloud, allowing the later retrieval of this data without loss of information. In order to demonstrate the efficiency of the ideas that guide the strategy proposed in this thesis, several experiments were carried out. To perform these experiments, a private cloud infrastructure managed by OpenStack (Openstack Cloud Operating System) was used. The algorithms that compose the QSM-EXTRACTION strategy were implemented in C++ language. In order implement the evaluation of the efficiency of the QSM-EXTRACTION strategy, a collection of syntactically created documents of different sizes was used. The results of the experiments proved the feasibility of using the proposed approach in scenarios typical of cloud