Multi-Agent Approach to Modeling and Implementing Fault-Tolerance in Reactive Autonomic Systems
Recently, autonomic computing has been proposed as a promising solution for software complexity in IT industry. As an autonomic approach, the Reactive Autonomic Systems Framework (RASF) proposes a formal modeling based on mathematical category theory, which addresses the self-* properties of reactiv...
| Main Author: | |
|---|---|
| Format: | Others |
| Published: |
2011
|
| Online Access: | http://spectrum.library.concordia.ca/7340/1/ShafieiDizaji_MASc_S2011.pdf Shafiei-Dizaji, Nassir <http://spectrum.library.concordia.ca/view/creators/Shafiei-Dizaji=3ANassir=3A=3A.html> (2011) Multi-Agent Approach to Modeling and Implementing Fault-Tolerance in Reactive Autonomic Systems. Masters thesis, Concordia University. |
| Summary: | Recently, autonomic computing has been proposed as a promising solution for software complexity in IT industry. As an autonomic approach, the Reactive Autonomic Systems Framework (RASF) proposes a formal modeling based on mathematical category theory, which addresses the self-* properties of reactive autonomic systems in a more abstract level.
This thesis is about the specification and implementation of the reactive autonomic systems (RAS) through multi-agent approach by laying emphasis on the fault-tolerance property of RAS. Furthermore, this thesis proposes a model-driven approach to transform the RAS model to agent templates in multi-agent model using Extensible Stylesheet Language Transformation (XSLT). The multi-agent approach in this research is implemented by Jadex, a high-level Java-based agent programming language. The intelligent agents are created in Jadex based on the Belief-Desire-Intension (BDI) agent architecture. The approach is illustrated on a case study.
|
|---|