Adaptive hierarchical and partitioned clustering for controlling high volume traffic data streams in peer to peer networks

Abstract

Peer to Peer P2P network traffic experiences a major portion of internet traffic and becomes the elementary data source while considering high volume data streams Among the different types of internet applications in today s scenario P2P network is one of the high traffic network characterized by which allows a large number of users to establish communication with each other which in turn directly access and download files from the peers machine and share the resources which are considered to be inevitable The basic requirement in the internet traffic is the classification of network to be performed for high volume data streams in P2P network In this research work the use of clustering techniques by dividing into hierarchical and partition clustering to identify interesting traffic patterns based on the volume of data from network traffic data in an efficient manner is categorized Earlier models dealt with clustering of data stream which handles network traffic that comprises of numerical data set by way of using K means algorithm whereas this proposed model works well for both numerical and categorical data set This research work has extended from numerical data set to categorical data set using K modes partition cluster algorithm for high volume data streams