移動式交易型代理人模式(mobile transactional agent model)會由ㄧ個任務代理人在不同的機器上依序執行不同階段的工作，當全部的階段執行完畢，才算完成此任務；否則任務就算失敗。傳統的方式，會因網路的不穩定(例如：擁塞和延遲)，都會使任務代理人無法準時到達，導致任務無法完成。本論文中提出一個可提高完成率和降低溝通時間的方法，利用同步訊息回覆的方法，找出數個可以繼續執行的下個階段，最快回應的一個階段(沒有產生網路擁塞的問題)，繼續執行。利用此方法，可以避免網路擁塞而沒有辦法繼續執行的問題，也可縮短階段間跟階段間，訊息溝通的時間。模擬的結果也驗證，我們的方法非常有效，可以大幅提高任務代理人的成功率和縮短階段間跟階段間訊息傳遞的時間。

A mobile transactional agent model divided a task into several stages and sequentially performed by a task agent at different machines. The task is finished only when all the stages have been performed. In the conventional method, the unstable conditions, such network congestion or delay will cause the task can not be finished. Accordingly, in this paper, we propose a method that can speed up the finish and decrease the communication among agents of a task. When a stages is finish, the method will choose some stages, which can be performed at next strategy, and then, send coordination messages to all of the targeted stages. The stages that first sends an acknowledgment back will be assigned as the next stages to be performed. Notably, any unstable network conditions will block the response time among agents. Therefore, our method can speed up the success rate of a task and reduce the communication time among agents. Simulation results also demonstrated our method is efficient.

摘要 I
誌謝 III
目錄 V
第1章 前言 1
1.1 背景及動機 1
第2章 相關背景及研究 4
2.1 主從式運算架構及移動式代理人運算架構的比較 4
2.2 代理人的概念 8
2.3 移動式代理人的概念 9
2.4 傳統交易型代理人 10
2.5 移動式交易型代理人 12
2.6 移動式代理人容錯方法 13
第3章 系統架構 17
第4章 系統策略 20
4.1 低延遲策略 20
4.2 檢查點策略 25
第5章 系統模擬與效能分析 28
5.1 模擬環境及參數 29
5.2 平均延遲時間 31
5.3 網路不穩定下的模擬 34
5.4 網路與軟體不穩定下的模擬 37
第6章 結論 41
圖 2 – 1 主從式運算架構 5
圖 2– 2 移動式代理人運算架構 6
圖 2 - 3 傳統式交易型代理人旅遊路程 10
圖 2 – 4 傳統交易型代理人相關性 11
圖 2 – 5 複製容錯機制 14
圖 2 – 6 檢查點容錯機制 16
圖 3 – 1 交易任務的階段關係 18
圖 3 – 2 階段排程 19
圖 4 – 1 流程圖 21
圖 4 – 2 交易任務的階段關係 23
表 5 – 1 模擬環境的參數定義 28
圖 5 – 1 平均延遲時間關係 32
圖 5 – 2 c=0.4的任務完成率 35
圖 5 – 3 c=0.2的任務完成率 35
圖 5 – 4 c=0.1的任務完成率 36
圖 5 – 5 c=0.4的任務完成率 37
圖 5 – 6 c=0.2的任務完成率 38
圖 5 – 7 c=0.1的任務完成率 38

[1]A.R. Naghsh Nilchi, A. Vafaei, H. Hamidi, "Evaluation of Security and Fault Tolerance in Mobile Agents," Wireless and Optical Communications Networks, 2008, pp. 1-5.
[2]A. Holt, C.-Y. Huang, J. Monk, "Performance analysis of mobile agents," Evaluation of Security and Fault Tolerance in Mobile Agents, vol. 1, 2007, pp. 532-538.
[3]Jin Yang, Jiannong Cao, and Weigang Wu, "CIC: An Integrated Approach to Checkpointing in Mobile Agent Systems," Semantics, Knowledge and Grid, 2006, pp. 4.
[4]K. Mohammadi and H. Hamidi, "An Approach to Fault-Tolerant Mobile Agent Execution in Distributed Systems," The First IEEE and IFIP International Conference in Central Asia, 2005, pp. 26-29.
[5]M.A.M. Ibrahim, "Distributed Network Management with Secured Mobile Agent Support," Hybrid Information Technology, 2006, vol. 1, pp. 244-251.
[6]M.R. Lyu, Xinyu Chen, and Tsz Yeung Wong, "Design and Evaluation of a Fault-Tolerant Mobile-Agent System," IEEE Digital Object Identifier, vol. 19, 2004, pp. 32-38.
[7]M. Silva, V. Batista, and J.G. Silva, "Fault-Tolerant Execution of Mobile Agents," Proc. Int''l Conf. Dependable Systems and Networks, IEEE CS Press, 2000, pp. 135-143.
[8]M. Dalmeijer, E. Rietjens, D. Hammer, A. Aerts, and M. Soede, "A Reliable Mobile Agents Architecture," Object-Oriented Real-Time Distributed Computing, 1998, pp. 64-72.
[9]P Pop, V. Izosimov, P Eles, Zebo Peng, "Design Optimization of Time- and Cost-Constrained Fault-Tolerant Embedded Systems With Checkpointing and Replication," IEEE Transactions, Very Large Scale Integration (VLSI) Systems, March 2009, Volume 17, pp. 389 - 402.
[10]Pathak H, Nipu, Garg K, "A Fault Tolerant Comparison Internet Shopping System: BestDeal by Using Mobile Agent," Information Management and Engineering, 2009. ICIME ''09. International Conference on, 2009 ,pp. 541-545.
[11]R.O. Abel, S. Dasgupta, and J.G. Kuhl, "Performance modelling of a fault-tolerant agent-driven system," IEEE International Conference on Control and Automation, June 2007, pp. 1559-1564.
[12]Safi, E.; Moshovos, A.; Veneris, A, "On the Latency and Energy of Checkpointed Superscalar Register Alias Tables," IEEE Transactions, Very Large Scale Integration (VLSI) Systems, 2009, Volume PP, pp. 1-1.
[13]S. Pears, J. Xu, and C. Boldyreff, "Mobile Agent Fault Tolerance for Information Retrieval Applications: An Exception Handling Approach," Autonomous Decentralized Systems, IEEE CS Press, 2003, pp. 115-122.
[14]S. Pleisch and A. Schiper, "Fault-Tolerant Mobile Agent Execution," IEEE Trans. Computing, 2003, vol. 52, no. 2, pp. 209-222.
[15]T. Osman, W. Wagealla, and A. Bargiela, "An Approach to Rollback Recovery of Collaborating Mobile Agents," IEEE Trans. Systems, Man and Cybernetics, 2003, Part C, vol. 34, no. 1, pp. 48-57.
[16]Wanlong Li, Dayou Liu, Hui Zhao, "Fault-Tolerance Mechanism of Mobile Agent In Mobile Ad Hoc Networks," Wireless Communications, Networking and Mobile Computing, Oct 2008, pp.1-4.
[17]Wang-Xiao Yun, Ye-Jian Nan, Huang-Guo Jin, Yu-Mei He, "Improvement of Temporal-Replication Mechanism in Mobile Agent System Fault-Tolerant Model," Computational Intelligence and Security Workshops, 2007, Dec. 2007, pp. 530-533.
[18]Xuejun Meng, and Huanguo Zhang, "An Efficient Fault-Tolerant Scheme for Mobile Agent Execution," Systems and Control in Aerospace and Astronautics, 2006, pp. 19-21.
[19]Y. Tanaka, N. Hayashibara, T. Enokido, and M. Takizawa, "A Fault-Tolerant Transactional Agent Model on Distributed Object Systems," Advanced Information Networking and Applications, 2006, vol. 2, pp. 18-20.
[20]Y. Tanaka, T. Enokido, and M. Takizawa, "Design and implementation of fault-tolerant transactional agents for manipulating distributed objects," Proceedings of the 10th IEEE International Workshop on Object-Oriented Real-Time Dependable Systems, Feb. 2005, pp. 329-336.
[21]施威銘，網路概論，旗標出版股份有限公司，2008，第1-11頁。
[22]竇其仁、林志敏、林正敏，網路代理人，知城數位科技股份有限公司，2006，第4-2 – 4-10頁。