Please use this identifier to cite or link to this item: http://localhost:8081/xmlui/handle/123456789/9719
Title: SIMULATION STUDY OF LOOP PREVENTION ALGORITHMS IN MPLS NETWORK'S
Authors: Suchdev, Aashish
Keywords: ELECTRONICS AND COMPUTER ENGINEERING;LOOP PREVENTION ALGORITHMS;MPLS NETWORK;ATM CELLS
Issue Date: 2001
Abstract: Multi protocol label switching (MPLS) is aimed at optimizing network resource utilization, while also providing tragic-engineering capabilities to the network administrator. It is viewed as an important part of the core of the proposed next generation networks. Central to this concept, are the label switched paths (LSPs), which are used for transmission of labeled IP packets through the MPLS domains, much the same way ATM cells are transported across VCs. Since MPLS is aimed at optimizing network resource utilization, formation of loops in LSPs has a detrimental effect on this very objective. In this thesis, we have implemented two algorithms proposed to the IETF as loop prevention measures. These are Path Vector algorithm, where loops are prevented by using a path vector which contains the ids of all the LSRs (label switching routers — MPLS nodes) downstream of a particular node, and Colored Thread algorithm, where loop prevention is implemented by propagating a unique color corresponding to the particular LSP connection, downstream. The performance of the two algorithms has been compared on the basis of number of control messages exchanged and bits transmitted to establish a path and those exchanged when an existing LSP needs to be reconfigured due to a next hop loss at one of the LSRs. It has been shown that the colored thread algorithm fares better during path establishment in the downstream on-demand label distribution mode, while the path vector algorithm performs better in the downstream unsolicited mode. The comparison was done by simulating a network and then generating path set up and reconfiguration requests. The model was developed in C++ on Windows98 environment...
URI: http://hdl.handle.net/123456789/9719
Other Identifiers: M.Tech
Research Supervisor/ Guide: Garg, Kumkum
metadata.dc.type: M.Tech Dessertation
Appears in Collections:MASTERS' THESES (E & C)

Files in This Item:
File Description SizeFormat 
ECDG10447.pdf3.65 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.