Title: Traffic Engineering and Quality of Service in the Internet The Internet, born in 1969, has become a ¿giant¿ that is crucial for the whole world. With the increasing use of Internet for commercial purposes, more stringent applications have been deployed, like Voice over IP (VoIP), Video on Demand (VOD), and Virtual Private Networks (VPN). These services require specific Quality of Service (QoS) to be ensured or guaranteed by the network. Internet Service Providers (ISPs) are driven to rely on traffic engineering to better control the flow of IP packets, to provide a good service and to fulfill their network management goals. This thesis focuses on the following aspects of the broad problem of traffic engineering and QoS in the Internet: - Intradomain traffic engineering with an emphasis on resources usage. - How ISPs carry out their interdomain traffic engineering in reality. - How to enable end-to-end QoS across the Internet.@en

In this paper we present DeSiNe, a modular flow-level network simulator. DeSiNe is aimed at performance analysis and benchmarking of Quality of Service routing algorithms and traffic engineering extensions. Several well-known QoS routing algorithms and traffic engineering extensions have been implemented in DeSiNe. The flow-level nature provides scalability, such that large networks and heavy-traffic conditions are possible. In this paper, the functional and structural design of DeSiNe are presented and the usability and various features are illustrated by menas of several examples. The source code of DeSiNe is publicly available.@en

Routing policies are typically partitioned into a few classes that capture the most common practices in use today[1]. Unfortunately, it is known that the reality of routing policies [2] and peering relationships is far more complex than those few cleasse [1,3]. We take the next step of searching for the appropiate granularity at which policies should be modeled. For this purpose, we study how and where to configure pre-prefix policies in an AS-level model of the INernet, such that the selected paths in the model are consistent with those observed in BGP data from mulitple vantage points. By comparing business relationships with per-prefix filters, we investigate the role and limitations of business relationships as a model for policies. We observe that popular locations for filtering correspond to balleys where no path shoulb be propagated according to inferred business relationships. This result reinforces the validity of the valley-free property used fro business relationships interference. However, given the sometimes large path diversity ASs have, business relationships do not contain enough information to devide which path will be chosen as the best. To model how individual ASs choose their best paths, we introduce a new abstraction: next-hop atoms. Next-hop atoms capture the fifferent sets of neighboring ASs adn AD uses for its best routes. We show that a large fraction of next-hop atoms correspond to per-neighbor path choices. A non-negligible fraction of path choices, however, correspond to hot-potato routing and tie-breaking within the BGP decision process, very detailed aspects of Internet routing.@en