Authentication improvement and intrusion detection system generation for hadoop framework / Mohamed Aboelhaggag Hassany Morsy ; Supervised Magdy M. Said Elsoudani , Mohsen Mohamed Abdelmonem Tantawy

Thesis (Ph.D.) - Cairo University - Faculty of Engineering - Department of Electronics and Communication

A framework called Hadoop, which has many tools, had been adopted to process the Big Data. The existing authentication protocol for Hadoop is Kerberos, which has a vulnerability against weak passwords and is not scalable. In this thesis, we propose an authentication protocol proper for Hadoop to address some of these issues. We call it Hash Token Based Authentication(HTBA), a token-based authentication approach to achieve scalability. In this work, we implement the proposed HTBA, Kerberos, and Hash-chain protocols on the Deter lab environment to conduct performance analysis. On the other hand, the IDS is an essential security service in Hadoop security layers. The IDS methodology is either signature-based detection or anomaly behavior detection. The use of DL to produce a model for the IDS may take a long time because of computation complexity and a large number of hyperparameters. We use Apache Spark, which is one of the Hadoop tools for this purpose. Different DL models for IDS on Apache Spark have been implemented in this thesis.We use the famous Network Security Lab - Knowledge Discovery and Data Mining (NSL-KDD) dataset and calculate a computation delay in Apache Spark. Moreover, an enhanced model has been proposed to improve attack detection accuracy

Issued also as CD