You are here

Trafic engineering techniques for data center networks

Traffic engineering (TE) consists in improving the performance of the telecomunication networks which is evaluated by a large number of criteria. The ultimate objective is to avoid congestion in the network by keeping its links from being overloaded. In large Ethernet networks, such as data centers, improving the performance of the traditional switching protocols is a crucial but very challenging task due to an exploration in the size of solution space and the complexity. Hence, exact methods are inappropriate even for reasonable size networks.

Local Search (LS) is a powerful method for solving computational optimization problems. The advantage of LS for these problems is its ability to find an intelligent path from a low quality solution to a high quality one in a huge search space. Thus, we propose different approximate methods based on LS for solving the class of TE problems in data center networks (DCN) that implement Spanning Tree Protocol and Multiple Spanning Tree Protocol.

In this thesis, we first tackle the minimization of: the maximal link utilization in the Ethernet networks with one or many spanning trees, the service disruption and the worst-case maximal link utilization in DCNs with many spanning trees. We then develop a novel design of multi-objective algorithms for solving the TE problems in large data centers by taking into account three objectives to be minimized: maximal link utilization, network total load and number of used links.

Our schemes reduce significantly the size of the search space by releasing the dependence of the solutions from the link cost computation in order to obtain an intended spanning tree. Our approaches show good results on the credible data sets and are evaluated by the strong assessment methods.