An efficient forward and backward fault-tolerant mobile agent system

ASIAIR > College of Computer Science > Department of Computer Science and Information Engineering > Proceedings > Item 310904400/8859

Please use this identifier to cite or link to this item: http://asiair.asia.edu.tw/ir/handle/310904400/8859

Title:	An efficient forward and backward fault-tolerant mobile agent system
Authors:	Chen, Jong-Shin;Shi-, Hong-Da;Chen, Ching-Mu;Hong, Zeng-Wen;Zhong, Pei-Lin
Contributors:	Department of Computer Science and Information Engineering
Keywords:	Computer system recovery;Fault detection;Intelligent systems;Mobile agents;Systems analysis;Failure Detection;Failure detection and recoveries;Failure Recovery;Fault-tolerant;Hardware faults;Mobile agent system;OR-networks;Present stage;Program code;System Fault-Tolerant;Task agents;Task process
Date:	2008
Issue Date:	2010-04-08 12:22:37 (UTC+0)
Publisher:	Asia University
Abstract:	Mobile agent is a special program and it can switch and execute the task of user ommands among the networks and hosts. During the task executing, the mobile agent can convey the data, state and program code to another host in order to autonomously execute and continue the task in another host. While the mobile agent is executing the task at any of the software and hardware fault incurred or network problems, there will be two conditions as following lists: 1. Users continuously wait the reply from the agent, but users will never have the reply because of the some faults incurred to the agent in the networks or hosts. 2. Users assign a new agent as the former agent has been lost to restart the former task, but the former agent only congested the delay problem of the network or host. This causes that theses two agents have the same task to be executed. Therefore, the fault detecting and recovering of the mobile agent are important issues to be discussed. However, this paper propose a Front Behind Failure Detection and Recovery method that the task agent has to report the task process at the present stage to the former and latter agent and agents will exchange their messages for the present stage. This is more accurate than the method in for the task at present because it can reduce the loading of the task fault report from the network congested. © 2008 IEEE.
Relation:	Proceedings - 8th International Conference on Intelligent Systems Design and Applications, ISDA 2008 :61-66
Appears in Collections:	[Department of Computer Science and Information Engineering] Proceedings

Files in This Item:

File	Description	Size	Format
		0Kb	Unknown	563	View/Open
250.doc		31Kb	Microsoft Word	248	View/Open

Loading...