Donald Bren School of Information and Computer Sciences

Individual-based simulation modeling structures an entire complex and open-ended application as a collection of individual autonomous entities. The behavior of the entire system is simulated as interactions among such entities and hence the application development focuses primarily on describing the entities behaviors. Therefore, this modeling scheme attracts many scientists dealing with complex models from the battlefield to particle-level simulations. A natural implementation of individual-based modeling is to use the philosophy of autonomous objects, i.e., mobile entities navigating autonomously through their underlying computational network. We have developed MESSENGERS, an autonomous-objects-based system aimed at a general-purpose distributed computing, and, especially, distributed simulations. It is the first system to provide a virtual-time computing environment for autonomous objects, and therefore the present work is the first experiment in applying the paradigm of autonomous objects to distributed individual-based simulations. The environment supports both conservative and optimistic simulations. In this paper, we discuss MESSENGERS' advantages for application development from the software engineering point of view, and show conditions, such as problem size, entity granularity and scalability, necessary for MESSENGERS to exploit its parallelism and achieve competitiveness with conventional message-passing executions.