Spatial Partitioning of a Battlefield Parallel Discrete-Event Simulation

Abstract

This thesis describes a method for spatially partitioning a battlefield into units known as sectors to achieve speedup two ways: through the reduction of each battlefield object's next event search space, and lowering the amount of message-passing required. Each sector is responsible for tracking and controlling access to all objects within its boundaries. A distributed proximity detection algorithm employing boundary-crossing events is used to control player movement between sectors. Each object's state information is replicated in all sectors it has sensor capability for the minimum time required; -this ensures that each object's next event is properly determined based upon interactions with objects in other sectors as well as its own. Each scenario is initialized using three sources of information: a set of scenario input files, a mapping file, and command-line arguments. Scenarios generate output in the form of screen messages, log files, and graphics display files. The issues involved in determining when and how to dynamically change the boundaries are discussed. A heuristic for changing sector boundaries based upon the number of players in each sector, as well as player attributes, is proposed.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Dec 01, 1992
Accession Number
ADA258911

Entities

People

  • Kenneth C. Bergman

Organizations

  • Air Force Institute of Technology

Tags

Communities of Interest

  • Air Platforms
  • Ground and Sea Platforms
  • Materials and Manufacturing Processes
  • Sensors
  • Weapons Technologies

DTIC Thesaurus Topics

  • Air Force
  • Algorithms
  • Computer Programming
  • Computer Programs
  • Computer Simulations
  • Computers
  • Department Of Defense
  • Dynamic Loads
  • Engineering
  • Hash Tables
  • Lists (Data Structures)
  • Operating Systems
  • Parallel Computing
  • Reliability
  • Simulators
  • Software Development
  • United States

Readers

  • Computer Vision.
  • Economics
  • Parallel and Distributed Computing.

Technology Areas

  • Space
  • Space - Space Objects