Type: Thesis (Masters)
Internet Traffic Engineering:
An Artificial Intelligence Approach.
Masters thesis, University of Porto.
This is the latest version of this eprint.
Full text not available from this archive.
Due to the diversification and increasing popularity of services provided over the Internet limitations of current mechanisms for Internet management are becoming evident. In particular, it is increasingly clear that the Internet network does not provide sufficient support for the efficient control and management of traffic, i.e. for
This dissertation addresses the problem of traffic engineering on the Internet from an Artificial Intelligence perspective. It argues that Internet traffic engineering should be automated instead of being performed reactively as response to undesirable network states. It presents and evaluates mechanisms for supporting Internet traffic engineering based on Artificial Intelligence methods.
This dissertation also discusses a suitable architecture for the application of the proposed mechanisms. It argues that the proposed mechanisms are able to support differentiated services useful for a wide range of applications. Finally, in a public heterogeneous network with the size of the Internet special attention should be given to the deployment of the proposed solutions. Consequently, the deployment of this architecture is analysed and the conclusion drawn that there are a number of feasible approaches toward deployment.
The work here presented argues the following: firstly, it is possible to implement mechanisms over the Internet framework that extend current protocols and algorithms enabling traffic engineering to be performed by operators; secondly, that applying these mechanisms reduces human intervention together with the management problems faced by operators and at the same time the efficiency with which the network can be run is improved; thirdly, that these improvements can correspond to increased network performance as viewed by the user; and finally, that not only the resulting deployment
but also the deployment process itself are feasible.
Available Versions of this Item
Archive Staff Only: edit this record