An Agent Based Distributed Computing System using XReggie and Ronin

December 1, 1999 - December 1, 2001

To realize the dream of truely ubiquitous information superhighway, it is necessary for information to be available from mobile platforms. But along with that, with the increasing variation in the capability of the different devices around us, efficient distribution of load is becoming very important to support ubiquitous computation. This project aims at building an agent-based middleware addressing the different issues of distributed computing in a mobile environment.

The project concentrates in developing a distributed Component based computation system using Ronin: a distributed Agent development framework and XReggie: An XML based service matching system. The main goal of the project is to realize the inherent limitations in a mobile computing environment and trying to address them. The system is developed on a distributed agent based infrastructure and introduces the concept of distributed computation in an agent scenario. The main idea behind the project is to effeciently utilize the resources available to a mobile device in its vicinity, whether it is in the wired network or whether the mobile device is in a adhoc bluetooth type network. The project develops a middleware to effeci\ ently discover services even without the knowledge of the discovery protocol being used, and manage the different software components required to properly execute a task in such a distributed agent scenario. This project uses XReggie to address the problem of interface matching to find services and uses XML based semantic matching primitives to discover services dynamically. The Agent infrastructure takes care of task distribution and management. It also introduces a uniform communication scheme between heteroggenous agents which further helps in the integration and interoperation of such agents. The system can be easily extended to plug in different heterogenous agents and hence provides the facility of extending the system to incorporate more intelligence and make more intelligent decisions.

OWL Tweet