Quantifying Interest Partitioning in a Distributed Model
			    David R. Pratt
		    Department of Computer Science
		      Naval Postgraduate School
		    Monterey, California, U. S. A.

 
As the large scale modeling and simulation community transitions from
monolithic models towards distributed models, there is an accompanying
need for a shift in the processing model used by model developers and
exercise designers. Interest management and data distribution management
schemes have the potential to reduce computer processing and networking
load enhancing the flexibility and scalability of models. This
presentation discusses the tradeoffs involved in implementing these
schemes and provides an empirical treatment of the computational aspects
of interest partitioning, a precursor to effectively using interest
management and data distribution management schemes among distributed
systems.



Dr. Pratt is an Associate Professor of Computer Science at the 
Naval Postgraduate School.  He joined this department full time
in 1993.  His research interests encompass all areas need to support
virtual worlds research, e.g. computer graphics, real-time programming,
3D simulation systems, virtual worlds systems, autonomous agent
simulation and development, user interfaces, parallel processing
operating systems, and terrain databases.  Since 1996 he has been the
Technical Director, Joint Simulation Systems (JSIMS) Joint Project
Office (JPO), Orlando Florida.