Show simple item record

dc.contributor.advisorVincent W. S. Chan.en_US
dc.contributor.authorCarey, Matthew F. (Matthew Francis)en_US
dc.contributor.otherMassachusetts Institute of Technology. Department of Electrical Engineering and Computer Science.en_US
dc.date.accessioned2016-07-18T20:03:56Z
dc.date.available2016-07-18T20:03:56Z
dc.date.copyright2016en_US
dc.date.issued2016en_US
dc.identifier.urihttp://hdl.handle.net/1721.1/103719
dc.descriptionThesis: Ph. D., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2016.en_US
dc.descriptionCataloged from PDF version of thesis.en_US
dc.descriptionIncludes bibliographical references (pages 355-362).en_US
dc.description.abstractWe present the fundamental framework for a multi-service, heterogeneous internetworking architecture that provides probabilistic, end-to-end quality of service guarantees to the data application for short critical messages prior to transmission. The class of critical network messages spans many applications in the civilian and defense network realms, including command and control of crucial public infrastructure, early warning alerts, and military command dissemination. These network applications share a need for a high probability of successful delivery with stringent delay requirements. The IP Internet is not well-suited to bear mission-critical messages across administrative domain boundaries with these demands because IP is a purely best effort network service and Border Gateway Protocol intentionally obscures the detailed administrative domain state information needed to describe the performance of internetwork paths. Current solutions to critical messaging involve the quasi-static provisioning of circuit connections, an approach that is not scalable for general mission requirements. The Critical Service architecture presented in this work directly addresses the need to generate a priori probabilistic guarantees without explicit path reservation overhead in an on-demand and dynamic fashion. The Critical Service architecture leverages a logically-centralized control plane, divorced from the network data plane, which aggregates a compact set of performance state information from participating networks. While these networks retain autonomy and routing control, we introduce the State Measurement Service necessary to characterize the intranetwork services offered by these networks to critical message datagrams. The learned performance measurements are aggregated and pruned by a local controller such that only the minimal required internal state is revealed to the centralized arbiter. Internetwork routing decisions for critical messages are made on a transaction-specific basis using this global state information and the specified service demands of the network application. An algorithmic method is developed that discovers internetwork paths and composes diversity-routed solutions that satisfy the requested minimum reliability and maximum tolerable delay bound performance requirements. A form of source routing using the computed internetwork service enforces the network-granularity routing solution.en_US
dc.description.statementofresponsibilityby Matthew F. Carey.en_US
dc.format.extent363 pagesen_US
dc.language.isoengen_US
dc.publisherMassachusetts Institute of Technologyen_US
dc.rightsM.I.T. theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. See provided URL for inquiries about permission.en_US
dc.rights.urihttp://dspace.mit.edu/handle/1721.1/7582en_US
dc.subjectElectrical Engineering and Computer Science.en_US
dc.titleCServ : an Internetwork architecture supporting mission-critical messaging with probabilistic performance guaranteesen_US
dc.title.alternativeInternetwork architecture supporting mission-critical messaging with probabilistic performance guaranteesen_US
dc.typeThesisen_US
dc.description.degreePh. D.en_US
dc.contributor.departmentMassachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
dc.identifier.oclc953415004en_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record