Skip navigation
Please use this identifier to cite or link to this item:
Title: Equivalence Class Snapshots in the Data Plane: A Measurement Framework for Network Analysis and Performance Error Diagnosis
Authors: Chang, Michael
Advisors: Rexford, Jennifer
Contributors: Narayana, Srinivas
Department: Computer Science
Class Year: 2016
Abstract: A network operator’s main responsibility is to maintain high performance in a network, a task greatly complicated by the complexity and enormous amounts of raw data that traverse most modern day networks. In response to the difficult of this task, many diagnosis tools have been created, each designed to address a different problem, and frequently require considerable start up time in terms of software and in some cases, hardware. In this paper, we propose a more general measurement framework that monitors traffic at the granularity of the Forwarding Equivalence Class (FEC). As an example of an application, we subsequently demonstrate that this measurement framework can be used to localize network congestion and diagnose the root case of this performance problem. We believe that such a framework can be extended to a wide variety of applications, such as performance benchmarking, network security, and forwarding correctness.
Extent: 56 pages
Type of Material: Princeton University Senior Theses
Language: en_US
Appears in Collections:Computer Science, 1988-2019

Files in This Item:
File SizeFormat 
Chang_Michael_thesis.pdf1.65 MBAdobe PDF    Request a copy

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