Show simple item record

dc.contributor.authorStoller, Scott D.en_US
dc.date.accessioned2007-04-23T18:09:32Z
dc.date.available2007-04-23T18:09:32Z
dc.date.issued1997-05en_US
dc.identifier.citationhttp://techreports.library.cornell.edu:8081/Dienst/UI/1.0/Display/cul.cs/TR97-1628en_US
dc.identifier.urihttps://hdl.handle.net/1813/7283
dc.description.abstractAs computers are integrated into systems that have stringent fault-tolerance requirements, there is a growing need for techniques to establish that these systems actually satisfy those requirements. Informal arguments do not supply the desired level of assurance for critical systems. This dissertation presents a rigorous, automated approach to analyzing distributed systems, with a focus on checking fault-tolerance requirements, and describes a prototype implementation of the analysis. The analysis is a novel hybrid of ideas from stream-processing semantics of networks of processes, abstract interpretation of programs, and symbolic computation. The underlying principles of the analysis method are general, but specialized techniques---such as the use of perturbations to represent changes to normal behavior caused by failures---are developed to deal efficiently with the types of systems and requirements that arise in establishing fault-tolerance. The method is illustrated with three examples: the Oral Messages algorithm for Byzantine Agreement, due to Lamport, Shostak and Pease, a standard protocol for FIFO reliable broadcast, and a (subtly) flawed protocol for fault-tolerant moving agents.en_US
dc.format.extent1978637 bytes
dc.format.extent2035483 bytes
dc.format.mimetypeapplication/pdf
dc.format.mimetypeapplication/postscript
dc.language.isoen_USen_US
dc.publisherCornell Universityen_US
dc.subjectcomputer scienceen_US
dc.subjecttechnical reporten_US
dc.titleA Method and Tool for Analyzing Fault-Tolerance in Systemsen_US
dc.typetechnical reporten_US


Files in this item

Thumbnail
Thumbnail

This item appears in the following Collection(s)

Show simple item record

Statistics