Adaptive agent allocation for massively multi-agent applications

Myeong Wuk Jang, Gui Agha

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

Abstract

Although distributed computing is necessary to execute massively multi-agent applications, the distribution of agents is challenging especially when the communication patterns among agents are continuously changing. This paper proposes two adaptive agent allocation mechanisms for massively multi-agent applications: one mechanism aims at minimizing agent communication cost, while the other mechanism attempts to prevent overloaded computer nodes from negatively affecting overall performance. We synthesize these two mechanisms in a multi-agent framework called Adaptive Actor Architecture (AAA). In AAA, each agent platform monitors the workload of its computer node and the communication patterns of agents executing on it. An agent platform periodically reallocates agents according to their communication localities. When an agent platform is overloaded, the platform migrates a set of agents, which have more intra-group communication than inter-group or inter-node communication, to a lightly loaded agent platform. These adaptive agent allocation mechanisms are developed as fully distributed algorithms, and they move the selected agents as a group. In order to evaluate these mechanisms, preliminary experimental results with large-scale micro UAV (Unmanned Aerial Vehicle) simulations are described.

Original languageEnglish (US)
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages25-39
Number of pages15
StatePublished - Dec 1 2005
Event1st International Workshop on Massively Multi-Agent Systems, MMAS 2004 - Kyoto, Japan
Duration: Dec 10 2004Dec 11 2004

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume3446 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other1st International Workshop on Massively Multi-Agent Systems, MMAS 2004
CountryJapan
CityKyoto
Period12/10/0412/11/04

Fingerprint

Communication
Vertex of a graph
Group Communication
Agent Architecture
Distributed computer systems
Unmanned aerial vehicles (UAV)
Communication Cost
Parallel algorithms
Distributed Algorithms
Distributed Computing
Locality
Workload
Monitor
Necessary
Evaluate
Experimental Results
Costs
Simulation
Actors

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Jang, M. W., & Agha, G. (2005). Adaptive agent allocation for massively multi-agent applications. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (pp. 25-39). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 3446 LNAI).

Adaptive agent allocation for massively multi-agent applications. / Jang, Myeong Wuk; Agha, Gui.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). 2005. p. 25-39 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 3446 LNAI).

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

Jang, MW & Agha, G 2005, Adaptive agent allocation for massively multi-agent applications. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 3446 LNAI, pp. 25-39, 1st International Workshop on Massively Multi-Agent Systems, MMAS 2004, Kyoto, Japan, 12/10/04.
Jang MW, Agha G. Adaptive agent allocation for massively multi-agent applications. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). 2005. p. 25-39. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
Jang, Myeong Wuk ; Agha, Gui. / Adaptive agent allocation for massively multi-agent applications. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). 2005. pp. 25-39 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{3923d806dfcb4a10be058fd1b734b328,
title = "Adaptive agent allocation for massively multi-agent applications",
abstract = "Although distributed computing is necessary to execute massively multi-agent applications, the distribution of agents is challenging especially when the communication patterns among agents are continuously changing. This paper proposes two adaptive agent allocation mechanisms for massively multi-agent applications: one mechanism aims at minimizing agent communication cost, while the other mechanism attempts to prevent overloaded computer nodes from negatively affecting overall performance. We synthesize these two mechanisms in a multi-agent framework called Adaptive Actor Architecture (AAA). In AAA, each agent platform monitors the workload of its computer node and the communication patterns of agents executing on it. An agent platform periodically reallocates agents according to their communication localities. When an agent platform is overloaded, the platform migrates a set of agents, which have more intra-group communication than inter-group or inter-node communication, to a lightly loaded agent platform. These adaptive agent allocation mechanisms are developed as fully distributed algorithms, and they move the selected agents as a group. In order to evaluate these mechanisms, preliminary experimental results with large-scale micro UAV (Unmanned Aerial Vehicle) simulations are described.",
author = "Jang, {Myeong Wuk} and Gui Agha",
year = "2005",
month = "12",
day = "1",
language = "English (US)",
isbn = "3540269746",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "25--39",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - Adaptive agent allocation for massively multi-agent applications

AU - Jang, Myeong Wuk

AU - Agha, Gui

PY - 2005/12/1

Y1 - 2005/12/1

N2 - Although distributed computing is necessary to execute massively multi-agent applications, the distribution of agents is challenging especially when the communication patterns among agents are continuously changing. This paper proposes two adaptive agent allocation mechanisms for massively multi-agent applications: one mechanism aims at minimizing agent communication cost, while the other mechanism attempts to prevent overloaded computer nodes from negatively affecting overall performance. We synthesize these two mechanisms in a multi-agent framework called Adaptive Actor Architecture (AAA). In AAA, each agent platform monitors the workload of its computer node and the communication patterns of agents executing on it. An agent platform periodically reallocates agents according to their communication localities. When an agent platform is overloaded, the platform migrates a set of agents, which have more intra-group communication than inter-group or inter-node communication, to a lightly loaded agent platform. These adaptive agent allocation mechanisms are developed as fully distributed algorithms, and they move the selected agents as a group. In order to evaluate these mechanisms, preliminary experimental results with large-scale micro UAV (Unmanned Aerial Vehicle) simulations are described.

AB - Although distributed computing is necessary to execute massively multi-agent applications, the distribution of agents is challenging especially when the communication patterns among agents are continuously changing. This paper proposes two adaptive agent allocation mechanisms for massively multi-agent applications: one mechanism aims at minimizing agent communication cost, while the other mechanism attempts to prevent overloaded computer nodes from negatively affecting overall performance. We synthesize these two mechanisms in a multi-agent framework called Adaptive Actor Architecture (AAA). In AAA, each agent platform monitors the workload of its computer node and the communication patterns of agents executing on it. An agent platform periodically reallocates agents according to their communication localities. When an agent platform is overloaded, the platform migrates a set of agents, which have more intra-group communication than inter-group or inter-node communication, to a lightly loaded agent platform. These adaptive agent allocation mechanisms are developed as fully distributed algorithms, and they move the selected agents as a group. In order to evaluate these mechanisms, preliminary experimental results with large-scale micro UAV (Unmanned Aerial Vehicle) simulations are described.

UR - http://www.scopus.com/inward/record.url?scp=26844461543&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=26844461543&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:26844461543

SN - 3540269746

SN - 9783540269748

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

SP - 25

EP - 39

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

ER -