Abstract: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. In this paper, we focus on using SDN as an approach for energy-aware routing (EAR). 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 table of OpenFlow switch can hold an infinite number of rules. In practice, this assumption does not hold since the flow table is implemented in Ternary Content Addressable Memory (TCAM) which is expensive and powerhungry. In this paper, we propose 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 present an exact formulation using Integer Linear Program (ILP) and introduce efficient greedy heuristic algorithm for large networks. Based on simulations, we show that using this smart rule space allocation, it is possible to save almost as much power consumption as in the classical EAR approach.

everl studies exhiit tht the tr0 lod of the routers only hs smll in)uE ene on their energy onsumptionF reneD the power onsumption in networks is strongly relted to the numer of tive network elementsD suh s interfesD line rdsD se hsE sisDFFF he gol thus is to (nd routing tht minimizes the @weightedA numer of tive network elements used when routingF sn this pperD we onsider simpli(ed rhiteture where onnetion etween two routers is represented s link joining two network interE fesF hen onnetion is not usedD oth network interfes n e turned o'F hereforeD in order to redue power onsumptionD the gol is to (nd the routing tht minimizes the numer of used links while stisfying ll the demndsF e (rst de(ne formlly the prolem nd we model it s n integer liner progrmF henD we prove tht this prolem is not in eD tht is there is no polynomilEtime onstntEftor pproximtion lgorithmF e propose heuristi lgorithm for this prolem nd we lso prove some negtive results out si greedy nd proilisti lgorithmsF hus we present study on spei( topologiesD suh s treesD grids nd omplete grphsD tht provide ounds nd results useful for rel topologiesF e then exhiit the gin in terms of numer of network interfes @leding to glol redution of pproximtely QQ wh for mediumEsized kone networkA for set of existing network topologiesX we see tht for lmost ll topologies more thn one third of the network interfes n e spred for usul rnges of opertionF pinllyD we disuss the impt of energy e0ient routing on the streth ftor nd on fult tolerneF Key-words: power onsumptionD energyEe0ient routingD grphsD integer liner progrmE mingD lgorithmsF