Adaptive approach for real-time data monitoring using mobile agents

Abstract

In this project, the performance of using mobile agent will be studied and assessed for real-time data monitoring in a distributed environment through the design and development of a prototype real-time data monitoring system, namely CoMAgent Proxy Model. The IBM Aglet, which is extensively used in research area, is adopted as the mobile agent components of the system. Mobile client is frequently disconnected from the network. Also, the bandwidth between mobile client and server is narrow. Using a client/server network computing paradigm [1] is insufficient and ineffective in real-time data monitoring, in particular for mobile client. This is the main problem of how to minimize the total communication and computational overheads for query evaluation and real-time data monitoring in a distributed environment. To address these issues, an adaptive Cooperative Mobile Agent Proxy Model (CoMAgent Proxy Model) is proposed. It is an integrated system by using mobile agent embedded intelligence to handle query in real-time data monitoring. The design of this model consists of three execution approaches including Parallel Execution Approach, Itinerary Execution Approach and Dynamic Execution Approach. Combining three execution approaches are to efficiently handle the system environment which the execution status is changing dynamically. Each approach handles the query evaluation by mixing the stationary agents and the mobile agents. Through the experiments, we demonstrated the CoMAgent Proxy Model that is effectively minimized the communication and computational overheads in real-time data monitoring.