Chapter

Towards Reliable Large-Scale Multi-agent Systems

DOI: 10.1007/11559221_43
Source: DBLP

ABSTRACT In this paper, we propose an approach for fault-tolerance of multi-agent systems (MASs). The starting idea is the application
of replication strategies to agents, the most critical agents being replicated to prevent failures. As criticality of agents
may evolve during the course of computation and problem solving, and as resources are bounded, we need to dynamically and
automatically adapt the number of replicas of agents, in order to maximize their reliability and availability. We will describe
our approach and related mechanisms for evaluating the criticality of a given agent and for deciding how to parameterize the
strategy (e.g., number of replicas). We also will report on experiments conducted with our prototype architecture (named DimaX).

Download full-text

Full-text

Available from: Zahia Guessoum, Feb 11, 2014
0 Followers
 · 
77 Views
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Consider a distributed internet-based market place with sellers and buyers represented by autonomous mobile agents. Individual users of the system may create agents to purchase or offer goods or services; these agents may travel over the network searching for bargains or for potential markets. What framework would best support such a system? An opportunistic user would always selfishly seek the best deal, which may translate into having access to as much information as possible (in a managable form, of course). In a distributed environment, this would mean spawning a very large number of agents, possibly organized as a tree, to disperse over the network. In a more volatile market needing local decisions, seller and buyer agents would want to be omnipresent. In both cases, agents serving the same interests will often need some type of coordination; an ability to coordinate the behavior of agents in agent ensembles is a key challenge for Distributed AI. From the perspective of a node hosting such activity, there has to be some incentive to allow it. Malicious or erroneous agents may threaten the node
    IEEE Intelligent Systems 04/1999; DOI:10.1109/5254.757630
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The authors describe how they extended a framework of active objects, named Actalk, into a generic multiagent platform, named DIMA. They discuss how they implemented this extension and report on one DIMA application that simulates economic models
    IEEE Concurrency 08/1999; DOI:10.1109/4434.788781
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Static organizational structures are not always suited for large-scale open multi-agent systems. On the other hand, emergent organizational structures may lead to undesirable behavior despite agent-level adaptation. We thus propose a new adaptive multi-agent architecture with both agent-level and organization-level adaptation. The global-level adaptation is based on the monitoring of the system's behavior and the dynamic reification of an organizational structure. This structure is used to prevent or detect undesirable behavior and take the required corrective actions. This architecture is applied to fault-tolerant multi-agent systems.