NetVolution research track

  • End-to-end Wireless Path Deployment with Intelligent Surfaces Using Interpretable Neural Networks
    C Liaskos, S Nie, A Tsioliaridou, A Pitsillides, S Ioannidis, I Akyildiz
    IEEE Transactions on Communications

NetVolution track


This research track constitutes my main research objective, at the TNL Lab, Institute of Computer Science, Foundation for Research and Technology - Hellas, Greece. 


Introduction: Evolvability of Internet Routing

The Internet is an enormous ecosystem of tens of thousands of independent domains that coordinate in an entirely distributed fashion through the Internet routing system. A major challenge that often surprises is that it is not possible to innovate in its core components, like the Internet Protocol (IP) and the Internet routing system. These components were designed almost 40 years ago and since then they have not changed much.

The Objective

The objective of the NetVolution project, led by prf. X. Dimitropoulos, is to foster innovation in the Internet routing system. NetVolution will exploit Software Defined Networking (SDN), a paradigm shift that is happening during the last few years in the area of computer networks that makes networks more programmable.

More about the project at the official web site:

The Envisioned Key-Contribution

"A platform to initiate and foster the migration of Internet routing to an open, evolvable paradigm.” 

Studied System Aspects

Three research aspects are presently studied: i) Providing robust, inter-domain Quality of Service, which is presently impossible, ii) Tackling recent Denial of Service variants that can even segment the Internet, but cannot be mitigated in the present routing status quo, iii) Minimizing the Internet routing convergence time, which has been observed to last excessively long, causing multiple problems in the Internet connectivity.

Personal contributions follow (see the full project publications here):

Inter-domain QoS

  • Fostering non-intrusive, inter-AS collaboration [[[Liaskos2015lightweight]]] [[[Liaskos.2013j3]]].

Security

  • Defense against Crossfire DDoS attacks [[[gkounis2016interplay]]] [[[Liaskos2016novel]]] [[[Liaskos.2013k3]]] [[[Liaskos.2013k3.1]]] [[[Liaskos.2013j3.1]]].

Conclusion

The current architecture of Internet can be considered as "ossified", i.e., it restrics innovation while yielding inability to meet contemporary needs in QoS and Security. E.g.:

  • The Crossfire attack can segment the Internet, but its mitigation requires inter-AS collaboration. This is presently not possible.
  • End-to-End QoS quarantees are not supported.

The NetVolution project aims at addressing these (and more) issues, using SDN principles.

Funding

This research is financed by the European Research Council (ERC) Starting Grant with a total funding of 1.41 million euros. (Contract Number: EU338402).

References

Powered by CMSimple| Template: ge-webdesign.de| Login