基于移动Agent的分布式资源发现研究与设计
|
摘要
移动Agent是一种可以在主机之间根据自主控制进行有目的移动的软件程序，其实质是一个封装了代码、运行状态和数据的智能计算实体。作为对象技术的更高层次的发展，移动Agent具有自主性、协作性、移动性和自适应性特点，能提供离线的计算模式，减少网络的通信流量，异步运行。它为解决移动环境中的动态资源发现问题提供了新的思路。
本文介绍了现有网络中的资源发现算法和协议，分析了传统资源发现算法的不足，针对移动网络高度动态性的特点，在泛洪算法的基础上，提出了一种基于移动Agent的分布式资源发现算法，设计并实现了基于此算法的资源发现系统。通过对系统运行情况的分析，进一步改进了该系统，提出了层次化的域模型概念；同时为提高移动Agent在移动网络中资源发现的灵活性，给出了几种不同的移动策略，根据移动策略动态修改移动Agent的旅行计划，使之能够更好地适应移动网络的动态性。
Mobile Agent is a kind of software which can make purposive movement among hosts by self control. It is, in essential, a compute entity which encapsulates codes, running states and data. As a higher development of object-oriented technology, mobile Agent has the characteristics of autonomy, good cooperation, mobility and self-adaptation, and it can make offline computing and reduce network traffic. This kind of technology contributes to the solution of dynamic resource discovery in the mobile environment. This thesis introduces the resource discovery algorithm of the current network, analyses the disadvantages of traditional resource discovery algorithms. Aimed at the highly dynamic characteristic of mobile network, this thesis puts forward an Agent-oriented distributed resource discovery algorithm, DDF(Double-Direction-Feedback algorithm) on the base of flooding algorithm. It also makes a design and implement of the resource discovery system which is based on the algorithm we proposed, and improves the system performance by bringing forward the concept of layered domain model. Meanwhile, to improve the flexibility of mobile Agent in the resource discovery, this thesis suggests several various travel policies, and revises Agent' s travel plan according to travel policies to make the Agent adapt to the dynamic characteristic of mobile network more better.
本文介绍了现有网络中的资源发现算法和协议，分析了传统资源发现算法的不足，针对移动网络高度动态性的特点，在泛洪算法的基础上，提出了一种基于移动Agent的分布式资源发现算法，设计并实现了基于此算法的资源发现系统。通过对系统运行情况的分析，进一步改进了该系统，提出了层次化的域模型概念；同时为提高移动Agent在移动网络中资源发现的灵活性，给出了几种不同的移动策略，根据移动策略动态修改移动Agent的旅行计划，使之能够更好地适应移动网络的动态性。
Mobile Agent is a kind of software which can make purposive movement among hosts by self control. It is, in essential, a compute entity which encapsulates codes, running states and data. As a higher development of object-oriented technology, mobile Agent has the characteristics of autonomy, good cooperation, mobility and self-adaptation, and it can make offline computing and reduce network traffic. This kind of technology contributes to the solution of dynamic resource discovery in the mobile environment. This thesis introduces the resource discovery algorithm of the current network, analyses the disadvantages of traditional resource discovery algorithms. Aimed at the highly dynamic characteristic of mobile network, this thesis puts forward an Agent-oriented distributed resource discovery algorithm, DDF(Double-Direction-Feedback algorithm) on the base of flooding algorithm. It also makes a design and implement of the resource discovery system which is based on the algorithm we proposed, and improves the system performance by bringing forward the concept of layered domain model. Meanwhile, to improve the flexibility of mobile Agent in the resource discovery, this thesis suggests several various travel policies, and revises Agent' s travel plan according to travel policies to make the Agent adapt to the dynamic characteristic of mobile network more better.
引文
1. Green, S. et al., "Software Agents: A review", Department of Computer Science, Trinity College Dublin, 1997
2. Danny B. Lange and Mitsuru Oshima," Seven Good Reasons for Mobile Agents", Communications of the ACM, Vol. 42, No. 3, MARDh 1999.
3. IBM Japan Aglets, http://www, trl. ibm. com/Aglets/index, html
4. Aglets Community, http://Aglets. sourceforge. net/
5. Aridor, Y. and D. B. Lange (1998), "Agent Design Patterns: Elements of Agent Application Design" The Second International Conference on Autonomous Agents (Agents'98), ACM Press.
6. Gnutella, http://www, gnutella, co. uk/
7. Cameron Ross Dunne, "Using Mobile Agents for Network Resource Discovery in Peer-to-Peer networks", http://www.acm.org/sigs/sigecom/exchanges/issue-2.3/Dunne.pdf,2000
8. Christian Bettstetter, Christoph Renner, "A comparison of service discovery protocols and implementation of the Service Location Protocol", Munich Institute of Communication Networks, 2000
9. Sun Microsystems, Inc., JXTA v1.0 Protocols Specification, Revision 1.1.1, 12/6/2001
10. Christian Bettstetter, Christoph Renner, "A comparison of service discovery protocols and implententation of the service location protocol", Sixth EUNICE Open European Summer School: Innovative Internet Applications, Twente, Netherlands, September 13-15, 2000.
11. Yaron Y. Goland, Ting Cai, Paul Leach, "Simple Service Discovery Protocol", IETF Draft draft-cai-ssdp-v1-03, txt, October 28, 1999.
12. Microsoft Corporation, "Universal Plug and Play Device Architecture Version 1.0", June 8, 2000.
13. John Rekesh ,"UPnP, Jini and Salutation-A look at some popular coordination frameworks for future networked devices", June 17,
1999.
14. Mor Harchol-Balter, Tom Leighton, and Daniel Lewin. Resource discovery in distributed networks. In 18th Annual ACMSIGACT/SIGOPS Symposium on Principles of Distributed Computing, May1999.
15. Ting Law and Kai-yeung Siu, "An O (log n) Randomized Resource Discovery Algorithm", In 14th Internatioanl Symposium on Distributed Computing, October 2000.
16. YungkooJun, Ladislau Boloni, Krzysztof Palacz, and Dan C, "Agent-Based Resource Discovery", October 8, 1999.
17. Shay Kutten and David Peleg,: "Deterministic distributed resource discovery", In 19th Annual ACM SIGACT/SIGOPS Symposium on Principles of Distributed Computing, July 2000.
18．王敏毅，面向移动环境的分布对象技术，博士论文，电子科技大学，2002．4
19．刘大有，杨鲲，陈建中，Agent研究现状与发展趋势，《软件学报》，11(3)：315-321，2000
20．王敏毅，周明天，移动计算技术现状与发展，《计算机应用》，18(9)：60-62，1998．9．
21．林碧英，曲俊华，吕洁，《网络系统中的计算模式》，《现代电力》，17(4)：11-21，2000．11
22．朱淼良，邱瑜，移动代理系统综述，《计算机研究与发展》，38(1)：16-21，2001．4
23．张松，阎保平，基于移动Agent的网络管理框架及实现，《计算机研究与发展》，36(8)：1007～1011，1999
2. Danny B. Lange and Mitsuru Oshima," Seven Good Reasons for Mobile Agents", Communications of the ACM, Vol. 42, No. 3, MARDh 1999.
3. IBM Japan Aglets, http://www, trl. ibm. com/Aglets/index, html
4. Aglets Community, http://Aglets. sourceforge. net/
5. Aridor, Y. and D. B. Lange (1998), "Agent Design Patterns: Elements of Agent Application Design" The Second International Conference on Autonomous Agents (Agents'98), ACM Press.
6. Gnutella, http://www, gnutella, co. uk/
7. Cameron Ross Dunne, "Using Mobile Agents for Network Resource Discovery in Peer-to-Peer networks", http://www.acm.org/sigs/sigecom/exchanges/issue-2.3/Dunne.pdf,2000
8. Christian Bettstetter, Christoph Renner, "A comparison of service discovery protocols and implementation of the Service Location Protocol", Munich Institute of Communication Networks, 2000
9. Sun Microsystems, Inc., JXTA v1.0 Protocols Specification, Revision 1.1.1, 12/6/2001
10. Christian Bettstetter, Christoph Renner, "A comparison of service discovery protocols and implententation of the service location protocol", Sixth EUNICE Open European Summer School: Innovative Internet Applications, Twente, Netherlands, September 13-15, 2000.
11. Yaron Y. Goland, Ting Cai, Paul Leach, "Simple Service Discovery Protocol", IETF Draft draft-cai-ssdp-v1-03, txt, October 28, 1999.
12. Microsoft Corporation, "Universal Plug and Play Device Architecture Version 1.0", June 8, 2000.
13. John Rekesh ,"UPnP, Jini and Salutation-A look at some popular coordination frameworks for future networked devices", June 17,
1999.
14. Mor Harchol-Balter, Tom Leighton, and Daniel Lewin. Resource discovery in distributed networks. In 18th Annual ACMSIGACT/SIGOPS Symposium on Principles of Distributed Computing, May1999.
15. Ting Law and Kai-yeung Siu, "An O (log n) Randomized Resource Discovery Algorithm", In 14th Internatioanl Symposium on Distributed Computing, October 2000.
16. YungkooJun, Ladislau Boloni, Krzysztof Palacz, and Dan C, "Agent-Based Resource Discovery", October 8, 1999.
17. Shay Kutten and David Peleg,: "Deterministic distributed resource discovery", In 19th Annual ACM SIGACT/SIGOPS Symposium on Principles of Distributed Computing, July 2000.
18．王敏毅，面向移动环境的分布对象技术，博士论文，电子科技大学，2002．4
19．刘大有，杨鲲，陈建中，Agent研究现状与发展趋势，《软件学报》，11(3)：315-321，2000
20．王敏毅，周明天，移动计算技术现状与发展，《计算机应用》，18(9)：60-62，1998．9．
21．林碧英，曲俊华，吕洁，《网络系统中的计算模式》，《现代电力》，17(4)：11-21，2000．11
22．朱淼良，邱瑜，移动代理系统综述，《计算机研究与发展》，38(1)：16-21，2001．4
23．张松，阎保平，基于移动Agent的网络管理框架及实现，《计算机研究与发展》，36(8)：1007～1011，1999