Zaheer Khan Zaheer2.Khan@uwe.ac.uk
Professor in Computer Science
Decentralized architecture for fault tolerant multi agent system
Khan, Zaheer; Shahid, Salman; Ahmad, Farooq; Ali, Arshad; Suguri, Hiroki
Authors
Salman Shahid
Farooq Ahmad
Arshad Ali
Hiroki Suguri
Abstract
Multi Agent Systems (MAS) are expected to be involved in
futuristic technologies. Agents require some execution
environment in which they can publish their service
interfaces and can provide services to other agent. Such
execution environment is called Agent Platform (AP).
From a technical point of view any abnormal behavior of
platform can distress agents residing on that platform.
That’s why it is necessary to provide a suitable
architecture for the AP which should not only provide fault
tolerance but also scalability features. There also exist
some management components within the platform, which
provide services to application agents. All the agents
within MAS are managed by Agent Management System
(AMS) which is the mandatory supervisory authority of
any AP. To be more scalable, a single agent platform can
be distributed over several machines which not only
provides load balancing but also fault tolerance depending
upon the distributed architecture of the AP. In existing
systems, AMS is centralized i.e. it exists on one machine.
With centralized AMS, this infrastructure lacks fault
tolerance, which is a key feature of high assurance.
Absence of fault tolerance is the main reason for the small
number of deployments of MAS. Failure of AMS leads
towards abnormal behavior in the distributed platform.
This paper proposes Virtual Agent Cluster (VAC)
paradigm which strongly supports decentralized AMS to
achieve fault tolerance in distributed AP. VAC provides
fault tolerance by using separate communication layers
among different machines. Experiments show that it
improves performance, brings autonomy and supports
fault recovery along with load balancing in distributed AP.
Presentation Conference Type | Conference Paper (unpublished) |
---|---|
Conference Name | International Symposium on Autonomous Decentralized Systems |
Start Date | Apr 4, 2005 |
End Date | Apr 8, 2005 |
Acceptance Date | Apr 4, 2004 |
Publication Date | Apr 4, 2004 |
Peer Reviewed | Peer Reviewed |
Pages | 167-174 |
Keywords | fault tolerance, scalability, decentralization |
Public URL | https://uwe-repository.worktribe.com/output/1060945 |
Additional Information | Title of Conference or Conference Proceedings : ISADS 2005 |
You might also like
Towards cloud based smart cities data security and privacy management
(2014)
Presentation / Conference Contribution
Using space-based downstream services for urban management in smart cities
(2014)
Presentation / Conference Contribution
Towards cloud based big data analytics for smart future cities
(2015)
Journal Article
Big data in Smart Cities
(-0001)
Book Chapter
Downloadable Citations
About UWE Bristol Research Repository
Administrator e-mail: repository@uwe.ac.uk
This application uses the following open-source libraries:
SheetJS Community Edition
Apache License Version 2.0 (http://www.apache.org/licenses/)
PDF.js
Apache License Version 2.0 (http://www.apache.org/licenses/)
Font Awesome
SIL OFL 1.1 (http://scripts.sil.org/OFL)
MIT License (http://opensource.org/licenses/mit-license.html)
CC BY 3.0 ( http://creativecommons.org/licenses/by/3.0/)
Powered by Worktribe © 2024
Advanced Search