Application of a massively multi-agent system to internet routing management

Osamu Akashi, Kensuke Fukuda, Satoshi Kurihara, Toshio Hirotsu, Toshiharu Sugawara

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Abstract

    Diagnosing the anomalies of inter-AS (autonomous system) routing and flexibly controlling its behavior to adapt to environmental changes are difficult, because this information changes spatially and temporally over different administrative domains. Multi-agent-based systems that coincide with this control architecture have been applied to these domains, but the number of deployed agents is small and more accurate analysis taking into consideration the actual Internet structure is desirable. To enable better analysis, cooperation among tens of thousands of agents is needed. This paper proposes a cooperative routing management system, called NetManager-M, which enables detailed analysis by using massively deployed agents on the Internet. NetManager-M can diagnose the routing flows around suspicious areas through cooperation among dynamically organized agents. This cooperation, which is conducted based on the current and previous routing topology, enables monitoring of routing update messages at neighboring observation points and the identification of the causes of problems in more detail. This system is thus an effective means of detailed analysis for typical scenarios of inter-AS routing failures.

    Original languageEnglish
    Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Pages131-145
    Number of pages15
    Volume5043 LNAI
    DOIs
    Publication statusPublished - 2008
    Event1st International Workshop on Coordination and Control in Massively Multi-agent Systems, CCMMS 2007 - Honolulu, HI
    Duration: 2007 May 152007 May 15

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume5043 LNAI
    ISSN (Print)03029743
    ISSN (Electronic)16113349

    Other

    Other1st International Workshop on Coordination and Control in Massively Multi-agent Systems, CCMMS 2007
    CityHonolulu, HI
    Period07/5/1507/5/15

    Fingerprint

    Multi agent systems
    Multi-agent Systems
    Routing
    Internet
    Agent-based Systems
    Topology
    Autonomous Systems
    Anomaly
    Monitoring
    Update
    Scenarios

    ASJC Scopus subject areas

    • Computer Science(all)
    • Theoretical Computer Science

    Cite this

    Akashi, O., Fukuda, K., Kurihara, S., Hirotsu, T., & Sugawara, T. (2008). Application of a massively multi-agent system to internet routing management. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 5043 LNAI, pp. 131-145). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5043 LNAI). https://doi.org/10.1007/978-3-540-85449-4_10

    Application of a massively multi-agent system to internet routing management. / Akashi, Osamu; Fukuda, Kensuke; Kurihara, Satoshi; Hirotsu, Toshio; Sugawara, Toshiharu.

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 5043 LNAI 2008. p. 131-145 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5043 LNAI).

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Akashi, O, Fukuda, K, Kurihara, S, Hirotsu, T & Sugawara, T 2008, Application of a massively multi-agent system to internet routing management. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 5043 LNAI, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 5043 LNAI, pp. 131-145, 1st International Workshop on Coordination and Control in Massively Multi-agent Systems, CCMMS 2007, Honolulu, HI, 07/5/15. https://doi.org/10.1007/978-3-540-85449-4_10
    Akashi O, Fukuda K, Kurihara S, Hirotsu T, Sugawara T. Application of a massively multi-agent system to internet routing management. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 5043 LNAI. 2008. p. 131-145. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-540-85449-4_10
    Akashi, Osamu ; Fukuda, Kensuke ; Kurihara, Satoshi ; Hirotsu, Toshio ; Sugawara, Toshiharu. / Application of a massively multi-agent system to internet routing management. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 5043 LNAI 2008. pp. 131-145 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
    @inproceedings{322460f20ea14433a90f8cfe2f12f5a2,
    title = "Application of a massively multi-agent system to internet routing management",
    abstract = "Diagnosing the anomalies of inter-AS (autonomous system) routing and flexibly controlling its behavior to adapt to environmental changes are difficult, because this information changes spatially and temporally over different administrative domains. Multi-agent-based systems that coincide with this control architecture have been applied to these domains, but the number of deployed agents is small and more accurate analysis taking into consideration the actual Internet structure is desirable. To enable better analysis, cooperation among tens of thousands of agents is needed. This paper proposes a cooperative routing management system, called NetManager-M, which enables detailed analysis by using massively deployed agents on the Internet. NetManager-M can diagnose the routing flows around suspicious areas through cooperation among dynamically organized agents. This cooperation, which is conducted based on the current and previous routing topology, enables monitoring of routing update messages at neighboring observation points and the identification of the causes of problems in more detail. This system is thus an effective means of detailed analysis for typical scenarios of inter-AS routing failures.",
    author = "Osamu Akashi and Kensuke Fukuda and Satoshi Kurihara and Toshio Hirotsu and Toshiharu Sugawara",
    year = "2008",
    doi = "10.1007/978-3-540-85449-4_10",
    language = "English",
    isbn = "3540854487",
    volume = "5043 LNAI",
    series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
    pages = "131--145",
    booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

    }

    TY - GEN

    T1 - Application of a massively multi-agent system to internet routing management

    AU - Akashi, Osamu

    AU - Fukuda, Kensuke

    AU - Kurihara, Satoshi

    AU - Hirotsu, Toshio

    AU - Sugawara, Toshiharu

    PY - 2008

    Y1 - 2008

    N2 - Diagnosing the anomalies of inter-AS (autonomous system) routing and flexibly controlling its behavior to adapt to environmental changes are difficult, because this information changes spatially and temporally over different administrative domains. Multi-agent-based systems that coincide with this control architecture have been applied to these domains, but the number of deployed agents is small and more accurate analysis taking into consideration the actual Internet structure is desirable. To enable better analysis, cooperation among tens of thousands of agents is needed. This paper proposes a cooperative routing management system, called NetManager-M, which enables detailed analysis by using massively deployed agents on the Internet. NetManager-M can diagnose the routing flows around suspicious areas through cooperation among dynamically organized agents. This cooperation, which is conducted based on the current and previous routing topology, enables monitoring of routing update messages at neighboring observation points and the identification of the causes of problems in more detail. This system is thus an effective means of detailed analysis for typical scenarios of inter-AS routing failures.

    AB - Diagnosing the anomalies of inter-AS (autonomous system) routing and flexibly controlling its behavior to adapt to environmental changes are difficult, because this information changes spatially and temporally over different administrative domains. Multi-agent-based systems that coincide with this control architecture have been applied to these domains, but the number of deployed agents is small and more accurate analysis taking into consideration the actual Internet structure is desirable. To enable better analysis, cooperation among tens of thousands of agents is needed. This paper proposes a cooperative routing management system, called NetManager-M, which enables detailed analysis by using massively deployed agents on the Internet. NetManager-M can diagnose the routing flows around suspicious areas through cooperation among dynamically organized agents. This cooperation, which is conducted based on the current and previous routing topology, enables monitoring of routing update messages at neighboring observation points and the identification of the causes of problems in more detail. This system is thus an effective means of detailed analysis for typical scenarios of inter-AS routing failures.

    UR - http://www.scopus.com/inward/record.url?scp=54249127356&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=54249127356&partnerID=8YFLogxK

    U2 - 10.1007/978-3-540-85449-4_10

    DO - 10.1007/978-3-540-85449-4_10

    M3 - Conference contribution

    AN - SCOPUS:54249127356

    SN - 3540854487

    SN - 9783540854487

    VL - 5043 LNAI

    T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

    SP - 131

    EP - 145

    BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

    ER -