aalto1 untyped-item.component.html
Large Scale Model Checking: Distributed State Space Generation using MapReduce
Loading...
URL
Journal Title
Journal ISSN
Volume Title
School of Science |
Master's thesis
Electronic archive copy is available via Aalto Thesis Database.
Checking the digitized thesis and permission for publishing
Instructions for the author
Instructions for the author
Authors
Date
Department
Major/Subject
Mcode
T-106
Degree programme
Language
en
Pages
ix + 73 s. + liitt. 5 s.
Series
Abstract
Parallel and distributed model checking has become a topic of growing interest since 1990s.
A distributed model checker can verify large models because it can have access to a large amount of memory and computing power.
Until now, there are many distributed model checkers that have been developed.
A novel distributed model checking design and its implementation are discussed in this Thesis.
This approach has two phases.
In the first phase, the swarm verification is performed to generate a partial state space.
In the second phase, the partial state space is used as input, and the remainder of the state space is generated by performing one level breadth-first search in a loop.
The implementation combines SPIN and Hadoop.
SPIN is a well-known model checker.
Hadoop is a well-known open source implementation of the MapReduce distributed computing framework.
Our design is independent of the model checker used, so it is possible to choose another model checker than SPIN.
The benchmark experiments proved the concept of our design is feasible.
Analysis on the experiment results revealed the strategy on how to achieve fast state space generation.
However, further improvements in the implementation are still needed to support additional features of SPIN in state space generation.