Bibliothèque de L'IMIST/CNRST

PH.D Université Mohammed V 2017

Cloud computing has been established as a popular computing paradigm in IT industry. The scale of data generated across the cloud computing platform has been growing massively since
its inception and rise to prominence. Machine-generated log data makes up a big part; it is generated at every layer in the cloud ecosystem, spanning a wide range of IT operations, from storage and computation to networking and application services. Analyzing log data enhances the security of the cloud computing platform.
Efficiently managing and analyzing cloud logs is a difficult and expensive task due the growth in size and variety of formats. In this work, we have identified the requirements for unlocking the valuable unstructured wealth of information residing in log data generated in the cloud, arguing in favor of using big data technologies to achieve those requirements.
We have also designed a methodology based on a binary-based approach for frequency mining correlated attacks in log data. This approach is designed to function using the MapReduce
programming model. Initial experimental results are presented and they serve as the subject of a data mining algorithm to help us predict the likelihood of correlated attacks taking place.