University of British Columbia
Fault tolerance for distributed explicit-state model checking
Abstract
dc:descriptionPReach, developed at the University of British Columbia and Intel, is a state of the art parallel model checker. However, like many model checkers, it faces reliability problems. A single crash causes the loss of all progress in checking a model. For computations that can take days, restarting from the beginning is a problem. To solve this, we have developed PReachDB, a modified version of PReach. PReachDB maintains the state of the model checking computation even across program crashes by storing key data structures in a database. PReachDB uses the Mnesia distributed database management system for Erlang. PReachDB replicates data to allow the continuation of the computation after a node failure. This project provides a proof-of-concept implementation with performance measurements.
Degree
thesis:*- Name thesis:degree_name
- Master of Science - MSc
- Level thesis:degree_level
- master's
- Discipline thesis:degree_discipline
- Computer Science
- Grantor dc:publisher
- University of British Columbia
- Year dc:date
- 2014
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Ishida, Valerie Lynn
Rights
dc:rights- Statement dc:rights
-
- Attribution-NoDerivs 2.5 Canada
- Language dc:language
- eng
Identifiers
dc:identifier.*- Handle dc:identifier
- http://hdl.handle.net/2429/50746