Project Description: |
Accurate identification of network traffic
according to the application type is an important task of network
management. For example, a network administrator may want to identify
and throttle/block traffic from peer to peer applications to manage
bandwidth budget and to ensure quality of service objectives are met
for business critical applications. Similar to network management
tasks, many network engineering problems such as workload
characterization, and modeling, capacity planning, traffic shaping/
policing and route provisioning also rely on accurate identification
of network traffic. However, an accurate method for reliably
identifying the applications associated with network traffic is still
to be developed. In this work, we are specifically interested in
automatically detecting/identifying ssh traffic using a hybrid
approach including graph and machine learning based techniques.
|