Adversarial Degradation of the Availability of Routing Infrastructures and Other Internet-Scale Distributed Systems
2016-06
View/Download File
Persistent link to this item
Statistics
View StatisticsJournal Title
Journal ISSN
Volume Title
Title
Adversarial Degradation of the Availability of Routing Infrastructures and Other Internet-Scale Distributed Systems
Authors
Published Date
2016-06
Publisher
Type
Thesis or Dissertation
Abstract
The Internet relies on its routing infrastructure, a globally spanning distributed system of special purpose computers call routers, to deliver packets between hosts. In order to build the paths data will travel, routers execute a routing protocol called the Border Gateway Protocol, or BGP. BGP is built to be decentralized and highly accommodating to arbitrary preferences of the organizations that own routers. This dissertation focuses on examining the following thesis statement. The current state of BGP, coupled with the Internet's extreme level of topological complexity, allows adversaries who can interact with BGP routers to degrade the availability properties of both the entire Internet routing infrastructure and other Internet-scale distributed systems. The research in this work breaks down into two independent arcs. The first arc focuses on attacks which aim to disrupt the availability of large portions on the Internet's routing infrastructure. Through both simulation and experimentation with representative devices, this work demonstrates that a variety of adversaries can prevent large portions of the Internet from being able to correctly build paths to end destinations. The second arc focuses on how those who control routers, and therefore can decide how the routers will select paths, can attack the availability of distributed systems which closely interact with the transit infrastructure of the Internet. Specifically the work shows how, by altering the BGP decision making process slightly, a variety of systems, ranging from censorship circumvention tools to surveillance systems, can be defeated by such an adversary without loss of general connectivity.
Keywords
Description
University of Minnesota Ph.D. dissertation.June 2016. Major: Computer Science. Advisor: Nicholas Hopper. 1 computer file (PDF); xiv, 145 pages.
Related to
Replaces
License
Collections
Series/Report Number
Funding information
Isbn identifier
Doi identifier
Previously Published Citation
Suggested citation
Content distributed via the University Digital Conservancy may be subject to additional license and use restrictions applied by the depositor. By using these files, users agree to the Terms of Use. Materials in the UDC may contain content that is disturbing and/or harmful. For more information, please see our statement on harmful content in digital repositories.