There are two possible broad ways for classification of network traffic.
1. Distributed (Volunteer Based Systems)
2. Centralized (Using Machine Learning Algorithms like C4.5, C5.0, DPI)
#WORK1
The Distributed approach is implemented as Volunteer based systems. It has VBS clients and VBS Servers. The VBS Client is installed oh user’s computer and it collects data and the application along with the name of it. The VBS servers collect these data sent by client in their database. VBS Client and server are designed such that they automatically get started with the operating system of the user computer and they run in background.
The VBS Client tracks the information of every packet that passes the network interface card. Packets the at are captured are classified into in flow…show more content… Thus in essence the VBS client has four important functions, each function executed as separate thread. The different functions are
• Packet capture
• Monitoring the sockets(ports)
• Generation and determining the nature of flow
• Transmission the data captured
The VBS server stores these statistical parameters to analyze the flow and classify it.
The VBS server also functions thread based. The binary file sent by the client is received by the server and for every new client separate directory is created and the corresponding binary file sent by the client is stored in it.
Three different threads does the job
Thread 1: Authenticates the client and gives an identifier to each
