Une approche de sécurité Big Data dans le Cloud Computing

Abstract

Nowadays, Big Data has reached every area of our lives because it covers many tasks in different operation. This new technique forces the cloud computing to use it as a layer, for this reason cloud technology embraces it as Big Data as a service (BDAAS). After solving the problem of storing huge volumes of information circulating on the Internet, remains to us how we can protect and ensure that this information are stored without loss or distortion. The aim of this thesis is to study the problem of safety in BDAAS; in particularly we will cover the problem of Intrusion detection system (IDS). In order to solve the problems tackled in the thesis, we have proposed a Self-Learning Autonomous Intrusion Detection system (SLA-IDS) which is based on the architecture of autonomic system to detect the anomaly data. In this approach, to add the autonomy aspect to the proposed system we have used mobile and situated agents. The implementation of this model has been provided to evaluate our system. The obtained findings show the effectiveness of our proposed model. We validate our proposition using Hadoop as Big Data Platform and CloudSim, machine-learning Weka with java to create model of detection.