The exponential growth of network users and their communication demands has led to a tangible increment of energy consumption in network infrastructures. A new networking paradigm called Software Defined Networking (SDN) was recently emerged, in which the packet forwarding data plane is decoupled from decision making control plane.
SDN simplifies network management by offering programmability of network devices, monitoring the real-time traffic rates and ability of fast rerouting. It also assists to lower link data rates via rate-adaptation technique which reduces considerably the power consumption of network. It is deemed that SDN opens new promising opportunities to improve network performance in general and energy efficiency in particular. In this paper, given current traffic load of an SDN based network, we exploit the rate adaptation techniques in SDN-enabled devices in order to reduce the energy consumption.
The main idea behind this thesis is to find a distribution of flows over pre-calculated paths which allows to adapt the transmission rate of maximum links into lower states. We first formulate the problem as an Integer Linear Programming (ILP) problem. Then, we present four different computationally efficient algorithms namely greedy first fit, greedy best fit, greedy worst fit and meta-heuristic Genetic Algorithm (GA) based method to solve the problem for a realistic network topology.
Simulation results show that the GA based method consistently outperforms the others proposed greedy algorithms and by applying this algorithm, between 32 % to 47 % of the energy can be saved depending on the size and density of the network topology.
| Date | 1 Jun 2017 |
|---|
| Original language | French |
|---|
| Awarding Institution | - École de technologie supérieure
|
|---|
| Supervisor | Mohamed Cheriet (Supervisor) |
|---|
Zemmouri, S. (Author),
Cheriet (Supervisor),
1 Jun 2017Student thesis: Master's thesis › Master in Engineering: Engineering