×
Gorodetskii, V. I.; Karsaev, O. V.; Samoilov, V. V.; Serebryakov, S. V.

Development tools for open agent networks. (English. Russian original) Zbl 1178.93012

J. Comput. Syst. Sci. Int. 47, No. 3, 429-446 (2008); translation from Izv. Ross. Akad. Nauk, Teor. Sist. Upr. 2008, No. 3, 106-124 (2008).
Summary: Distributed peer-to-peer computing in which no centralized control server is assumed, and in some cases, client-server architecture is actually rejected, has been the subject of intense investigations in the field of information technologies in the recent years. This new paradigm of designing large-scale distributed intelligent systems attracts attention of both researchers and developers of industrial applications. This paradigm provides qualitatively new capabilities, especially for designing systems consisting of a large number of autonomous instances which can appear and disappear from the system at any time instant. The idea of integration of these systems (hereinafter, P2P systems for brevity) and multi-agent technologies is especially attractive; it realizes the advantages of both paradigms, namely, the maturity of technologies and the relative simplicity of creation of multi-agent systems consisting of autonomous intelligent programs which interact via a high-level language, as well as the openness, the scalability, and the reliability of P2P systems. This integration seems quite natural. Indeed, multi-agent systems is the set of autonomous intelligent programs; therefore, the structuring of these programs into the network whose nodes interact directly without any intermediate server is the natural step if the technology of P2P interactions has been well developed using a high level language. However, for practical realization of this idea, it is necessary to make one very important step, namely, to develop mechanisms for agent interaction on a semantic level, when agents query some functionalities (services) from each other, possibly, not knowing a priori what particular agents can provide the queried functionality (service) and whether these agents exist in the network. The application of the agent variant of the architecture, the service oriented architecture, another new paradigm of construction of information systems, provides practical capability of realization of mechanisms of semantic interaction of agents of the open P2P network. This work represents the P2P agent platform developed by the authors whose instances installed on the network nodes above the standard P2P service form the distributed meta-knowledge knowledge base intended for organization of semantic P2P interaction of agents. Application-oriented agents placed on the network nodes above the instances of the agent platform receive the capability of semantic interaction. The program realization of the platform is supported by the mechanism of P2P interaction of agents using messages with standard formats and pair communications of the network nodes. The role, functions, and the essence of the processes of operation of this platform are explained using two applications as the examples; these applications are sufficiently important from the practical point of view. They are used for verification of basic solutions presented in this work.

Cited in 2 Documents

MSC:

93A14 Decentralized systems
93C83 Control/observation systems involving computers (process control, etc.)
90B18 Communication networks in operations research

Keywords:

decentralized control; distributed intelligent systems; large-scale distributed intelligent systems; P2P systems; agent interaction
PDFBibTeX XMLCite
\textit{V. I. Gorodetskii} et al., J. Comput. Syst. Sci. Int. 47, No. 3, 429--446 (2008; Zbl 1178.93012); translation from Izv. Ross. Akad. Nauk, Teor. Sist. Upr. 2008, No. 3, 106--124 (2008)
Full Text: DOI

References:

[1] S. Androutsellis-Treotokis and D. Spinellis, ”A Survey of Peer-to-Peer Content Distribution Technologies”, ACM Computing Surveys, 36(4), pp. 335–371 (2004). · doi:10.1145/1041680.1041681
[2] Peer-to-Peer Research Group, http://www.irtf.org/charter?gtype=rg&group=p2prg (June 10, 2007).
[3] A. Tate, ”The ”Helpful Environment”: Geographically Dispersed Intelligent Agents that Collaborate”, IEEE Intell. Syst., pp. 57–61 (May–June, 2006).
[4] Fipa Agent Message Transport Service Specification, http://www.fipa.org/specs/fipa00067/ (June 10, 2007).
[5] FIPA P2P Nomadic Agents Working Group (P2P NA WG 6), http://www.fipa.org/subgroups/P2PNAWG.htm (June 10, 2007).
[6] FIPA P2P NA WG 6 Functional Architecture Specification Draft 0.12, http://www.fipa.org/subgroups/P2PNAWG-docs/P2PNA-Spec-Draft0.12.doc (June 10, 2007).
[7] V. Gorodetsky, O. Rarsaev, V. Samoylov, et al., ”Agent Platform: Implementation and Testing”, in Proceedings of International Workshop on Agent and Peer-to-Peer Computing Associated with AAMAS-07, Honolulu, Hawaii, 2007, pp. 21–32.
[8] V. Gorodetsky, O. Rarsaev, V. Samoylov, et al., ”Agent-Based Service-Oriented Intelligent P2P Networks for Distributed Classification, in Proceedings International Conference ”Hybrid Information Technologies”, Cheju Island, Korea, 2006, pp. 224–233.
[9] V. Gorodetsky, O. Rarsaev, V. Samoylov, et al., ”Multi-Agent Peer-to-Peer Intrusion Detection”, accepted for publication in Proceedings of 4th International Conference on Mathematical Methods, Models and Architectures for Computer Network Security, St. Petersburg, Russia, 2007.
[10] G. Li, ”Why JXTA”, www.javable.com/columns/p2p/solutions/03/(June 10, 2007).
[11] N. Krishnan, ”JXTA Solutions for P2P”, www.javable.com/java-world/10_01/01/#
[12] ”Introduction to Windows Peer-to-Peer Networking”, www.Microsoft.com/technet/network/p2p/p2pintro.mspx (June 10, 2007).
[13] FIPA, http://www.fipa.org/ (June 10, 2007).
[14] ”Available Realizations of Agent Platforms According to FIPA Specifications”, http://www.fipa.org/resources/livesystems.html (June 10, 2007).
[15] P.-M. Ricordel and Y. Remazeau, ”From Analysis to Deplayment: a Multi-Agent Platform Survey”, http://www.lia.deis.unibo.it/confs/ESAW00/pdf/ESAW07.pdf (June 10, 2007).
[16] ”Evaluation of Agent Platforms”, http://iag.pg.gda.pl/iag/down-load/leszezyna04evaluation.pdf (June 10, 2007).
[17] G. Nguyen, T. Dang, L. Hluchy, et al., ”Agent Platform Evaluation and Comparison”, http://pellucid.ui.sav.sk/TR-2002-06.pdf (June 10, 2007).
[18] P. Kungas, M. Matskin, ”Semantic Web Service Composition through a P2P-Based Milti-Agent Environment”, in Series: Lecture Notes in Computer Science. Sublibrary: Lecture Notes in Artificial Intelligence, 4118, pp. 106–119, ed. by Z. Despotovic, S. Joseph, C. Sartori (Springer, 2006).
[19] M. Wang, H. Wolf, M. Purvis, et al., ”An Agent-Based Collaborative Framework for Mobile Peer-to-Peer Applications”, in Series: Lecture Notes in Computer Science. Sublibrary: Lecture Notes in Artificial Intelligence, 4118, pp. 132–144, ed. by Z. Despotovic, S. Joseph, C. Sartori (Springer, 2006).
[20] UPNP Forum, www.upnp.org (June 10, 2007).
[21] A. Carrillo-Ramos, J. Gensel, M. Villanova-Oliver, et al., ”A Peer Ubiquitous Multi-Agent Framework for Providing Nomadic Users with Adapted Information”, in Series: Lecture Notes in Computer Science. Sublibrary: Lecture Notes in Artificial Intelligence, 4118, pp. 159–172, ed. by Z. Despotovic, S. Joseph, C. Sartori (Springer, 2006).
[22] B. Yu and M. P. Singh, ”Incentive Mechanism for Agent-Based Peer-to-Peer Systems, in Series: Lecture Notes in Computer Science. Sublibrary: Lecture Notes in Artificial Intelligence, 4118, pp. 77–88, ed. by Z. Despotovic, S. Joseph, C. Sartori (Springer, 2006).
[23] H. Z. Zhang and V. Lesser, ”Multi-Agent Based Peer-to-Peer Information Retrieval Systems with Concurrent Search Sessions”, in Proceedings of the Fifth International Joint Conference on Autonomous Agents and Multi-Agent Systems, Hakodate, Japan, 2006, pp. 305–312.
[24] S. Abu-Draz and E. Shakshuki, ”Peer to Peer Multi-Agent System for Online Trading”, in Series: Lecture Notes in Computer Science. Sublibrary: Lecture Notes in Artificial Intelligence. 2671, pp. 611–613 ed. by Y. Xiang, C. Brahim (Springer, 2003).
[25] M. Panti, L. Penserini, L. Spalazzi, et al., ”A Multi-Agent System Based on the P2P Model to Information Integration”, http://sra.itc.it/people/penserini/documents/LNCS_proposal38.pdf (June 10, 2007). · Zbl 1023.68819
[26] D. Bertolini, P. Busetta, M. Nori, et al., ”Peer-to-Peer Multi-Agent Systems Technology for Knowledge Management Applications. An Agent-Oriented Analysis”, in Proceedings of the Workshop ”From Objects to Agents” in Information to Knowledge (Milan, Italy, 2002), pp. 1–6. · Zbl 1021.68560
[27] M. Purvis, M. Nowostawski, S. Cranefield, et al., ”Multi-Agent Interaction Technology for Peer-to-Peer Computing Environment”, in Series: Lecture Notes in Computer Science. Sublibrary: Lecture Notes in Artificial Intelligence. 2872, pp. 150–161 ed. by G. Moro, C. Sartori, M. Singh (Springer, 2004). · Zbl 1087.68658
[28] ”FIPA Agent Communication Language Specification”, http://www.fipa.org/specs/fipa00061 (June 10, 2007).
[29] ”FIPA Agent Management Specification”, http://www.fipa.org/specs/fipa00023 (June 10, 2007).
[30] N. Lin, K. Marzullo, and S. Masini, ”Gossip versus Deterministic Flooding: Low Message Overhead and High Reliability for Broadcasting on Small Networks”, http://citeseer.ist.psu.edu/563854.html (June 10, 2007).
[31] ”FIPA Agent Discovery Service Specification”, http://www.fipa.org/specs/fipa00095/ (June 10, 2007).
[32] Pacific Marine Environmental laboratory, http://www.pmel.noaa.gov/tao/ (June 10, 2007).
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.