Experimental Application of Non-Relational Database Technology for Scaled Simulation Based Training

The next generation of virtual training applications for the Army needs to support training at the company or battalion level with corresponding increases to the operational area and level of complexity. Current simulation-based virtual environment training applications are capped at 40 participants or less due to the inability of systems resources to handle a larger number of concurrent users. This also means that there are limited system resources left over to accommodate larger operational areas and realistic levels of complexity. This severely limits the operational relevance of virtual training for many training scenarios. This research will provide designs, performance benchmarks, and a prototype of a generalized non-relational database management system for a virtual environment training application. Such a database design will attempt to overcome weaknesses of traditional relational databases and allow for hundreds or thousands of concurrent participants while preserving resources for increased operational areas and complexity. The dynamic schema of non-relational databases will also provide benefits such as faster application development, more reliable code integration, and continuous availability.