ABSTRACT:
There are two parts of this presentation:
1. Long term view: Here we want to talk about our Internet 3.0 ideas
coming from GENI/FIND type of clean slate research. The key concept we
want to emphasize in this part is the organizational representation in
the current Internet. This leads to several security and policy issues.
The first such issues appeared in the form of infrastructure ownership
(service provider domains) in late 80s and resulted in BGP routing and
subsequent issues that are still keeping RRG busy. The ownership issues
are now moving to hosts, humans, and data being which could belong to
different organizations. This results in a three tier virtualization of
the Internet. Routing in this environment would be quite different and
sub-optimal in the same way as BGP routing is due to clouds in the
infrastructure layer. The key issues that decide the route are policies
of the cloud owners. Path cost as modeled in traditional networking are
secondary.
2. Short Term View: Given the infrastructure clouds of today, we have
the problems of inter-domain routing, which is becoming complex and
unscalable. RRG has already identified the problems of Routing
Scalability, Traffic Engineering, Mobility, Multihoming, Renumbering,
Decoupling Location and Identification, Routing Quality, Routing
Security, and Incremental Deployability. A number of solutions have
been proposed, such as, LISP-ALT, LISP-NERD, APT, IVIP, TRRP, CRIO,
SIX/One routers, HIP, Shim6, I3, and Hi3. Most of these solutions
address one or other problem. We are developing solutions that try to
address all these aspects and more. In this part of the talk we will
give a brief overview of this approach.
Presentation slides in Adobe Acrobat Format: 1 slide/page
View Slides+Audio
Right click and save to download
Slides+Audio podcast.
Frequently asked questions about remote audio/video viewing