Distributed Simulation Systems

The choice of a proper hardware and software environment is very important in the design of distributed simulation systems. Conventional simulators, because of their centralized scheduling, do not fully utilize parallelism in computer systems. A new approach to simulation systems based on decentralized scheduling is studied in this report. In this approach, there is no central entity responsible for scheduling of tasks, but scheduling is distributed over all the processing units. A concurrent computer programming language is needed to program distributed simulation systems. Various concurrent prograrrrning languages are available. To select the proper language for distributed simulation, an understanding of these languages, especially the features useful in implementing distributed scheduling, is essential. This paper compares Ada and OCCAM for use as concurrent programming languages for distributed simulation. These languages are useful in distributed processing environments because of their tasking and inter-process communication features