Evaluation of IDS using Neural Network over Cloud

In this modern world evaluation of cloud computing has come forward for combination of logical entities like data and software which are accessible through internet. Innovations are necessary to ride the inevitable tide of change. Most of organizations are discord to decreases their computing cost through the means of virtualization. This demand of reducing the cost has led to the moderns of Cloud Computing. Cloud Computing offers better computing through improved exploitive and diminished administration and infrastructure costs. Cloud Computing is the sum of Software as a Service and subservience Computing. Cloud Computing is still at its infant stage and a very new technology for the enterprises. An ID is a vital component to maintain network security. Also, as the cloud platform is speedily evolving and become most popular in our day to day life, it is helpful and necessary to build effective IDS for the cloud computing. However, existing IDS will be likely to face challenges when deployed on the cloud platform. The predestined Intrusion Detection System architecture may lead to overburden of a part of the cloud due to the extra detection overhead. This thesis proposes a neural network based Intrusion Detection System that is a distributed system with an adaptive architecture, so as to make full use of the available resources without overloading any single machine in the cloud. Moreover, with the machine learning ability from the neural network, the proposed IDS can detect new types of attacks with fairly exact results. Evaluation of the introduced IDS with the KDD dataset on a physical cloud tested shows that it is a liberal of promises approach to detecting attacks in the cloud computing infrastructure.