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 publicationMassively Multi-Agent Technology - AAMAS Workshops - MMAS 2006, LSMAS 2006, and CCMMS 2007, Hakodate, Japan, May 9, 2006, Honolulu, HI, USA, May 15, 2007, Selected and Revised Papers
Pages131-145
Number of pages15
DOIs
Publication statusPublished - 2008 Oct 27
Externally publishedYes
Event1st International Workshop on Coordination and Control in Massively Multi-agent Systems, CCMMS 2007 - Honolulu, HI, United States
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)0302-9743
ISSN (Electronic)1611-3349

Other

Other1st International Workshop on Coordination and Control in Massively Multi-agent Systems, CCMMS 2007
CountryUnited States
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

  • Theoretical Computer Science
  • Computer Science(all)

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 Massively Multi-Agent Technology - AAMAS Workshops - MMAS 2006, LSMAS 2006, and CCMMS 2007, Hakodate, Japan, May 9, 2006, Honolulu, HI, USA, May 15, 2007, Selected and Revised Papers (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.

Massively Multi-Agent Technology - AAMAS Workshops - MMAS 2006, LSMAS 2006, and CCMMS 2007, Hakodate, Japan, May 9, 2006, Honolulu, HI, USA, May 15, 2007, Selected and Revised Papers. 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 Massively Multi-Agent Technology - AAMAS Workshops - MMAS 2006, LSMAS 2006, and CCMMS 2007, Hakodate, Japan, May 9, 2006, Honolulu, HI, USA, May 15, 2007, Selected and Revised Papers. 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, United States, 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 Massively Multi-Agent Technology - AAMAS Workshops - MMAS 2006, LSMAS 2006, and CCMMS 2007, Hakodate, Japan, May 9, 2006, Honolulu, HI, USA, May 15, 2007, Selected and Revised Papers. 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. Massively Multi-Agent Technology - AAMAS Workshops - MMAS 2006, LSMAS 2006, and CCMMS 2007, Hakodate, Japan, May 9, 2006, Honolulu, HI, USA, May 15, 2007, Selected and Revised Papers. 2008. pp. 131-145 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{a9e912b984f643649ffe058065e10487,
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",
month = "10",
day = "27",
doi = "10.1007/978-3-540-85449-4_10",
language = "English",
isbn = "3540854487",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "131--145",
booktitle = "Massively Multi-Agent Technology - AAMAS Workshops - MMAS 2006, LSMAS 2006, and CCMMS 2007, Hakodate, Japan, May 9, 2006, Honolulu, HI, USA, May 15, 2007, Selected and Revised Papers",

}

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 - 2008/10/27

Y1 - 2008/10/27

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

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

SP - 131

EP - 145

BT - Massively Multi-Agent Technology - AAMAS Workshops - MMAS 2006, LSMAS 2006, and CCMMS 2007, Hakodate, Japan, May 9, 2006, Honolulu, HI, USA, May 15, 2007, Selected and Revised Papers

ER -