13.07.14 - 16.07.14, Seminar 14291

Information-Centric Networking 3

Diese Seminarbeschreibung wurde vor dem Seminar auf unseren Webseiten veröffentlicht und bei der Einladung zum Seminar verwendet.

Motivation

Information­centric networking (ICN) is a communication paradigm where the basis for communication is named data. Instead of using endpoints to define communication (one device talking to another), ICN relies on asking for pieces of information by name independently of location.

Over the past few years, a large number of people have become involved in ICN research. Starting with the first Dagstuhl seminar on ICN in 2010, a number of high profile events have facilitated growth of a vibrant international research community.

People in both industry and academia have been involved in the various ICN project groups. They have designed architectures, created prototypes, built test beds and evaluated algorithms. So far we have learned that ICN is a truly transformational technology. ICNs are potentially more secure, more efficient and more functional than traditional networks. They can also exhibit a better incentive model than traditional networks.

We are now at a stage where several of these projects have working code ready to be tested in the real world. Various demos have shown the benefits and the potential that ICN can bring. Small-scale applications running over ICN have been demonstrated. But, we are still in our infancy when it comes to dealing with Internet­wide networking problems. If ICN is to be successful, we need to be able to scale.

At this seminar we will focus on two key aspects to ICN success; scalability and deployment.

Scalability in ICN has a number of unique problems. Switching over from a limited pool of hierarchical routable addresses to the limitless world of named objects stresses every network component. How can routing algorithms scale? How can packet processing and forwarding scale? How can management scale? How can memory scale?

Deployment of ICN systems faces a number of challenges. Not only do we want to deploy an incompatible network protocol, but we also want to do it as we redesign the APIs and we turn the incentives on their head. Overlays and islands will become the norm for a while. Without coordination and the glimpse of interoperability we will have a hard time driving adoption at an internet-wide scale.

Scalability and deployment are issues faced by all ICN architectures and we believe it is important to come together and talk about the potential hurdles we face and solutions for overcoming them.