Path Allocation in Backbone Networks: Project Report

Introduction There are two major frameworks for providing service guarantees in IP-based networks: The Integrated Services (IS) architecture and the Differentiated Services (DS) architecture. While the IS framework provides service guarantees using a flow-based packet differentiation, the DS architecture differentiates the treatment of aggregates. Though IS is able to provide strong Quality of Service (QoS) guarantees, its practical use suffers from significant scalability problems. The DS framework addresses this issue through its concept of aggregation. Packets are identified by simple markings that indicate the respective class. In the core of the network, routers do not need to determine to which flow a packet belongs, only which aggregate behavior has to be applied. Edge routers mark packets by setting the DS Code-Point (DSCP) and indicate whether they are within profile. If they are out of profile they might even be discarded by a dropper at the edge router. A particular marking on a packet indicates a so-called Per Hop Behavior (PHB) that has to be applied for forwarding the packet. The price for building aggregates is, however, a weaker control over the provisioned service parameters. The derivation of delay bounds remains a challenge, when providing a Premium Service. In [14] such bounds are derived for a general topology and a maximum load. However, these bounds can be improved, when additional information concerning the current load and the special topology of the DS domain is available. In [54], a central resource management for DS domains called Bandwidth Broker is presented. A Bandwidth Broker is a middleware service which controls and facilitates the dynamic access to network services of a particular administrative domain [31]. The task of a Bandwidth Broker in a DS domain is to perform a careful admission control, and to set up the appropriate configuration of the domain's edge routers, whereas the configuration of core routers remains static to allow for scalability. While doing so, the Bandwidth Broker knows about all service requests. Besides it can easily learn about the DS domains topology, either statically or by implementing a listener for the domains routing protocol. Thus, a Bandwidth Broker can have access to all information that is required, in order to base its admission control on delay bounds that are derived for the current load, and the special topology of the administrated domain. The particular challenge that arises here is to derive a set of formulas that can effectively be …