Software-defined Networks (SDN), in particular OpenFlow, is a new networking paradigm enabling innovation through network programmability. Over past few years, many applications have been built using SDN such as server load balancing, virtual-machine migration, traffic engineering and access control. For example, recently, Google has deployed during three years a network connecting their data centers across the planet using SDN. In this internship, we will focus on using SDN for energy-aware routing (EAR). Energy efficiency of networking infrastructure (telecom, data center, ISP or enterprise networks) is a growing concern, due to both increasing energy costs and worries about CO2 emissions. Since traffic load has a small influence on power consumption of routers, EAR allows to put unused links into sleep mode to save energy. SDN can collect traffic matrix and then computes routing solutions satisfying QoS while being minimal in energy consumption. However, prior works on EAR have assumed that the forwarding tables can hold an infinite number of rules. In practice, this assumption does not hold since these tables are implemented with Ternary Content Addressable Memory (TCAM) which is expensive and power-hungry.
In [1], we have proposed an optimization method to minimize energy consumption for a backbone network while respecting capacity constraints on links and rule space constraints on routers. In details, we have presented an exact formulation using Mixed Integer Linear Program (ILP) and introduced efficient greedy heuristic algorithm that were analyzed through simulations. The objectives of this internship are to extend this work in order to take into account the possibility to add wildcards within the forwarding rules. As an example, assuming that the rules are the mapping of [(src, dest): port-to-foward], we can compact several different rules with same source s, same port-to-forward p, but different destinations into: [(s,*): p]. In this case, the forwarding tables will be shrinked, but the general problem becomes more complicated. Therefore, the candidate will study the extension of the linear program to take this into account, study new heuristics, and perform the corresponding simulations on a realistic topology and set of traffic demands.
Prerequisites for candidates:
- Knowledge and/or taste for Networking
- Knowledge and/or taste for discrete mathematics, graph theory and/or combinatorial optimization
Contact information:
For further information, please contact
- Frédéric Giroire – frederic.giroire@cnrs.fr
- •Joanna Moulierac – joanna.moulierac@unice.fr
Location: COATI – INRIA/I3S (CNRS-UNS) – 2004 route des lucioles, 06906 SOPHIA ANTIPOLIS https://team.inria.fr/coati
References:
[1] Optimizing Rule Placement in Software-Defined Networks for Energy-aware Routing. Frédéric Giroire, Joanna Moulierac and Khoa Truong Phan. In Globecom 2014.
[2] On the effect of forwarding table size on SDN network utilization. R. Cohen, L. Lewin-Eytan, J. S. Naor, D. Raz In INFOCOM, 2014
[3] Software-Defined Networking: A Comprehensive Survey. Diego Kreutz, Fernando M. V. Ramos, Paulo Verissimo, Christian Esteve Rothenberg, Siamak Azodolmolky, Steve Uhlig In arXiv:1406.0440
URL : https://team.inria.fr/coati
URL Sujet detaille : www-sop.inria.fr/members/Joanna.Moulierac/SDN_Energy_internship_2014_COATI.pdf
