Skip to Main Content
Frequently Asked Questions
Submit an ETD
Global Search Box
Need Help?
Keyword Search
Participating Institutions
Advanced Search
School Logo
Files
File List
sargent_thesis.pdf (1.25 MB)
ETD Abstract Container
Abstract Header
On Understanding the Internet Via Edge Measurement
Author Info
Sargent, Matthew
ORCID® Identifier
http://orcid.org/0000-0002-6395-3787
Permalink:
http://rave.ohiolink.edu/etdc/view?acc_num=case1432290894
Abstract Details
Year and Degree
2015, Doctor of Philosophy, Case Western Reserve University, EECS - Computer and Information Sciences.
Abstract
The design philosophy of the Internet enables the transmission of packets between “smart” network edges via a “dumb” middle, or core portion of the network. Core networks are ultimately responsible for the single task of routing packets between hosts that are not physically connected. They do so by operating at the network layer of the Open System Interconnection model using the Internet Protocol. Routing requires no understanding of what type of traffic is being transmitted, but only of where a packet is ultimately destined. Whereas the core network’s primary job is to correctly route packets, edge networks have additional “smarts” as they contain myriad end point devices each responsible for implementing network protocols and applications. Common tasks like transmitting email, streaming video, or visiting web pages must all be implemented by end point devices. Thus, edge networks and the devices they contain are largely responsible for the evolution of overall network characteristics. In this dissertation we examine several changes edge networks have undergone recently and leverage empirical measurements to understand how edge network evolution has affected various network characteristics. We begin by studying traffic characterization and connection performance on a residential Fiber-To-The-Home network. We then shift our attention to the Transmission Control Protocol (TCP) and how its performance has been affected by a mismatch between the protocol specification and packet round trip times on the modern network. Next, we examine ways network applications drive sending patterns in TCP connections and how these patterns affect overall TCP performance. We conclude with two studies focusing on security related topics for edge networks. First, we develop and test a methodology that aims to broadly understand the port filtering policies in place throughout the network through passive traffic observation. Finally, we undertake a study of Internet Group Management Protocol (IGMP) traffic characteristics and demonstrate how IGMP traffic can be used to launch several attacks from edge networks. The list of topics we tackle is by no means exhaustive, but each topic does represent important work that allows us to keep our understanding of edge network behaviors up-to-date.
Committee
Mark Allman (Advisor)
Michael Rabinovich (Committee Chair)
Liberatore Vincenzo (Committee Member)
Gultekin Ozsoyoglu (Committee Member)
Francis Merat (Committee Member)
Marvin Schwartz (Committee Member)
Pages
137 p.
Subject Headings
Computer Science
Keywords
Computer Network Measurement
;
Protocol Performance
;
Fiber-To-The-Home Networks
Recommended Citations
Refworks
EndNote
RIS
Mendeley
Citations
Sargent, M. (2015).
On Understanding the Internet Via Edge Measurement
[Doctoral dissertation, Case Western Reserve University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=case1432290894
APA Style (7th edition)
Sargent, Matthew.
On Understanding the Internet Via Edge Measurement.
2015. Case Western Reserve University, Doctoral dissertation.
OhioLINK Electronic Theses and Dissertations Center
, http://rave.ohiolink.edu/etdc/view?acc_num=case1432290894.
MLA Style (8th edition)
Sargent, Matthew. "On Understanding the Internet Via Edge Measurement." Doctoral dissertation, Case Western Reserve University, 2015. http://rave.ohiolink.edu/etdc/view?acc_num=case1432290894
Chicago Manual of Style (17th edition)
Abstract Footer
Document number:
case1432290894
Download Count:
834
Copyright Info
© 2015, all rights reserved.
This open access ETD is published by Case Western Reserve University School of Graduate Studies and OhioLINK.