[attachment=7296]

---

Presented by:
Ghazi Bouabene
Christophe Jelger
Christian Tschudin
Stefan Schmid
Ariane Keller, and Martin May

The Autonomic Network Architecture (ANA)

ABSRACT
The objective of autonomic networking is to enable the autonomous formation and parametrization of nodes and networks by letting protocols sense and adapt to the networking environment at run time. Besides its dynamic aspects, a core requirement of autonomic networking is to define a structured framework and execution environment that enables algorithms to operate in a continously changing environment. This paper presents the major design principles of the Autonomic Network Architecture (ANA) and reports on a first implementation. The guiding principle of ANA is to strive for flexibility and genericity at all levels of the architecture. In our approach we explicitly avoid to impose a “one-size-fits-all” architecture (where communication protocols and paradigms are fixed by the architecture). To this end, ANA introduces generic abstractions, for example “information dispatch points” instead of addressable endpoints, as well as communication primitives that support network heterogeneity, adaptability, and evolution. These core abstractions allow for the coexistance of multiple and diverse networking styles and protocols. With the public release of the ANA prototype, we aim at federating autonomics related networking projects, enabling different actors to share, compare, and build upon each other’s work. The ANA runtime can host clean slate network designs as well as legacy Internet technology and serves as a platform for demonstrating autonomic communication principles.


INTRODUCTION
END USERS might have the impression that the Internet is working just fine, which usually is confirmed through our daily experience. When looking under the hood, however, one realizes that considerable manual configuration effort is necessary to keep this system running. In addition, a considerable number of operational and protocol patches (e.g. CIDR, NAT) had to be introduced in order to sustain its growth and increasing complexity. However despite these changes, the Internet core suite of protocols has almost not evolved since the 70ies, with new features such as IP multicast, IP mobility, and IP version 6 being slow or having failed to be widely deployed

A growing fraction of researchers and network operators are actually seriously concerned about the ability of the Internet to sustain more disruptive evolution steps. For many researchers including some of the early Internet designers the salient design principles (i.e. end-to-end principle and transparency, and global addressing) that contributed to the success of the Internet are also preventing any radical update of its core functionalities. As a result, the networking community has witnessed a renewed interest in the design of clean-slate network architectures (see e.g., [3], [4], [5], [6]) that could better support disruptive evolution and emerging network paradigms such as content-based, ad hoc, sensor, mobile, and delay-tolerant networking. A key motivation is to foster innovation by freeing researchers from providing backwards compatibility with the current Internet .
One direction of clean-slate research looking at adaptive and automated network operation is autonomic networking. The term “autonomic” (first coined in 2001 by IBM [7]) refers to the ability of a system to perform its operation via (so-called) self-* properties i.e., without requiring active human intervention. Current network protocols cover some part of the self-* properties, with for example routing and address autoconfiguration leading (to some extent) to the selforganisation and self-configuration of networks .
However, today these protocols are confined to narrow tasks (i.e. each protocol is designed in isolation to solve a specific problem) and they are not interlinked because of the exponential complexity of dealing with all possible interactions. If we want to develop more reliable and predictable communication systems, this disjoined development cycle must end and be replaced by a global and clearly structured development and execution framework. This would permit to sustain network evolution and at the same time reduce the overhead of human management, and it would mitigate the interaction complexity by allowing protocols and algorithms to operate in an autonomic manner
To support these objectives, the Autonomic Network Architecture (ANA) project [8] is exploring novel ways of organising and using networks beyond legacy Internet technology.We are designing and developping a network architecture that can demonstrate the feasibility and properties of autonomic networking. The main guiding principle behind the architectural work in ANA is to strive for a maximum degree of flexibility in order to support functional scaling by design at all levels of the architecture. Functional scaling means that a network is able to extend both horizontally (adding more functionality) as well as vertically (different ways of integrating abundant functionality).