Abstract
This paper proposes solutions based on mobile agents using mono-agent and multi-agent strategies for distributed meeting scheduling problem. Mobile agents are used in a typical decision-making scenario where they migrate to a calendar server to schedule a meeting between participants having time-conflicts. The mobile agent will either find a schedule matching all constraints or propose the best relaxed schedule violating some low-priority constraints, if no schedule respecting all constraints is found. The performance comparison is based on two metrics: the execution time and the total network load generated. Performance study results identify the cost of many operations used by the agent-based paradigm, such as inter-agent communications and agent migration, in relation to the nature and size of information exchanged. The overall experimental results provide a glimpse of the possibilities that software agents offer to solve distributed problems.
Keywords
Get full access to this article
View all access options for this article.