Content and Context Aware Networking Using Semantic Tagging

Abstract

Todays model of networking primarily concentrates intelligence at the end hosts with the network itself offering a simple``best-effort",``data agnostic" communication medium. However, such a paradigm has proven to be insufficient to meet todays needs considering the diversity of applications and devices that are networked. To offer value added services to these end users and applications, more and more intelligence needs to be migrated away from the edges and into the network in a controlled and tractable manner. In this paper, we present our approach of utilizing semantic data tagging to provide content level information for data streams flowing through a network. A policy based management mechanism is utilized within the network fabric allowing routers to reason over the content and make intelligent decisions regarding the handling of data packets. Service differentiation, in-network content adaptation, traffic monitoring and control etc. are some of the new services that can now be offered by the network in a generic and flexible manner. By deploying our proposed architecture, a network need no longer be viewed as a simple data transport medium but rather as a policy-controlled intelligent packet/stream processor that can offer specialized handling based on application needs.