Share Email Print

Proceedings Paper

Runtime services for sharing high-performance distributed data structures among HLA simulations
Author(s): Donald M. Leskiw; Junmei Zhau
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

This paper reports on results from an ongoing project to develop methods for representing and managing multiple, concurrent levels of modeling detail and enabling high performance computing, namely parallel processing, within object-based simulation frameworks such as HLA. We present here the interface structure and runtime support service concept for using parallel arrays for high performance computing within distributed object-based simulation frameworks. The approach employs a distributed array descriptor, which can be a basis for extending the HLA standard to provide support for efficiently sharing very large data arrays or sub-arrays among federates. The goal is to reduce communications overhead and thereby improve simulation performance involving C4ISR models that require, for example, interpolation and extrapolation of large data sets, such as those that naturally occur for overlay, coupling, and fusion of phenomenology information in multi- sensor networks.

Paper Details

Date Published: 23 June 2000
PDF: 8 pages
Proc. SPIE 4026, Enabling Technology for Simulation Science IV, (23 June 2000); doi: 10.1117/12.389364
Show Author Affiliations
Donald M. Leskiw, Ultra Corp. (United States)
Junmei Zhau, Ultra Corp. (United States)

Published in SPIE Proceedings Vol. 4026:
Enabling Technology for Simulation Science IV
Alex F. Sisti, Editor(s)

© SPIE. Terms of Use
Back to Top