A framework for modelling dependable real-time distributed systems
Authors:
Yeong-Jia Chen a;
Daniel Moss
a;
Shi-Kuo Chang a
a;
Shi-Kuo Chang a
| Affiliation: | a Department of Computer Science, University of Pittsburgh, Pittsburgh, PA, U.S.A. |
DOI:
10.1080/00207729708929466
Publication Frequency:
12 issues per year
Published in:
International Journal of Systems Science,
Volume
28,
Issue
11
July
1997
, pages 1025
- 1043
Subjects:
Artificial Intelligence;
Automation;
Automation Control;
Control Engineering;
Cybernetics;
Dynamical Control Systems;
Dynamical Systems;
Electronics;
Evolutionary Computing;
General Systems;
Intelligent Systems;
Networks;
Non-Linear Systems;
Statistics & Probability: Operations Research;
Industrial Engineering & Manufacturing: Operations Research;
Simulation & Modeling;
Supply Chain Management;
Systems & Control Engineering;
Systems & Controls;
Systems Architecture;
Systems Engineering;
Formats available:
PDF
(English)
Also incorporating: Systems Analysis Modelling Simulation
View Article:
View Article (PDF)
Abstract
A systematic way to introduce fault-tolerant and time-dependent properties into a complex software system is presented. The approach is based on an extended Petri Net model, called the G-Net, enhanced with a deterministic timing scheme. In the Timed G-Net (TGN) model, exception handling and timing mechanisms are used to introduce fault-tolerant and real-time properties. The basic idea is to use basic building blocks with timing properties to allow objects to treat a timing error by raising an exception and triggering some corrective actions. Using TGNs as these basic building blocks, we fur titer refine the model into a Smart Object (SMO) model. In each SMO, a TGN specification associated with a set of attributes, monitor/error-handler links and protocols support the construction of more reliable applications through an automatic transformer. This framework can contribute to alleviating the burden on the developers in designing a complex real-time dependable system. We use examples to show that a simplified real-time distributed air traffic control system can be modelled effectively using the SMO model.
|
| view references (20) |

Download Citation

CiteULike
Del.icio.us
BibSonomy
Connotea