In this section

    A Stochastic Approach To Solving The Weight Setting Problem in OSPF Networks

    (2007) A Stochastic Approach To Solving The Weight Setting Problem in OSPF Networks. Masters thesis, KFUPM.

    [img]
    Preview
    		 PDF (MS Thesis)
    Final_Thesis_Write_Up.pdf
    Download (1MB) | Preview

    Abstract

    In the world of Internetworks, to maintain a good connectivity of household, business and commercial computing, an exraordinary talent is important. Unpredictable dysfunction in its proper administration adds to the problems of this sophisticated network. One of the contributions in attempting to maintain the proper functioning of internetworking is made by the Open Shortest Path First (OSPF) protocol. It is a link state protocol designed to overcome the gap created by the Routing Information Protocol (RIP) in the internetworking domain. OSPF calculates the shortest paths from each source to all destinations using the Dijkstra’s algorithm based on the weights assigned to the links. In the past, various attempts have been made to resolve the congestion issues of Traffic Engineering. With such complex issues in the frameset, assigning weights to these large networks, resulting in the best cost is an NP-hard problem. In this thesis, a prudent approach of mitigating the mentioned problem by using a Stochastic Evolution (StocE) heuristic is used which provides a close to optimal solution to these kinds of problems. Through this work, an attempt has been made to optimize the weights on the network so as to minimize congestion. This approach is well supported by the results embedded towards the end of the work. Another core issue addressed in this work is the improvement of the network by considering single link failure scenarios. Two innovative strategies have been developed, where the same set of optimized weights for both topologies, i.e., with-link and without-link-failure, have been considered.

    Item Type: Thesis (Masters)
    Subjects: Computer
    Department: College of Computing and Mathematics > Computer Engineering
    Depositing User: S. M. REHMAN
    Date Deposited: 10 Mar 2008 11:35
    Last Modified: 30 Dec 2020 13:16
    URI: http://eprints.kfupm.edu.sa/id/eprint/264