Award Date

1-1-1999

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Electrical and Computer Engineering

First Committee Member

Shahram Latifi

Number of Pages

123

Abstract

The new generation of network services is being developed for incorporation in communication infrastructure. These services, generally called Quality of Services (QoS), should accommodate data file, video, and audio applications. The different performance requirements of these applications necessitate a re-examination of the main architectural components of today's networks, which were designed to support traditional data applications. Routing, which determines the sequence of network nodes a packet traverses between source and destination, is one such component. Here, we examine the potential routing problems in future Internet and discuss the advantages of class-based multi-path routing methods. The result is a new approach to routing in packet-switched networks, which is called Two-level Class-based Multipath routing with Prediction (TCMP). In TCMP, we compute multiple paths between each source and destination based on link propagation delay and bottleneck bandwidth. A leaky bucket is adopted in each router to monitor the bottleneck bandwidth on equal paths during the network's stable period, and to guide its traffic forwarDing The TCMP can avoid frequent flooding of routing information in a dynamic routing method; therefore, it can be applied to large network topologies.

Keywords

Computer Networks; Internet; Multipath Routing; Packet Switched Networks; Packet-switched Networks; Quality Of Service; Routing

Controlled Subject

Electrical engineering; Computer science

File Format

pdf

File Size

3491.84 KB

Degree Grantor

University of Nevada, Las Vegas

Language

English

Permissions

If you are the rightful copyright holder of this dissertation or thesis and wish to have the full text removed from Digital Scholarship@UNLV, please submit a request to digitalscholarship@unlv.edu and include clear identification of the work, preferably with URL.

Identifier

https://doi.org/10.25669/a9ch-baye


Share

COinS