University of Houston
Department of Computer Science
In partial fulfillment of the Requirements for the Degree of Master of Science

Osmund Castellino
will defend his thesis

Network parameter optimization and parallelization techniques for optimizing distributed visualization applications

Abstract

This study examines how a distributed visualization application can be optimized for communication speed and processing speed. In this study, we build a 3D volume rendering application as a test bed for performance measurements, using the OpenDX library, an open source development tool for visualization. We discuss the issues involved in the creating of a volume in OpenDX. We then run the application as a client-server program in order to test how TCP/IP network parameters can be tuned to reduce communication delays between visualization servers and clients. We use X-Windows to enable us to view the visualizations on the client terminal. Finally we study research in parallelization techniques that could apply to visualization. We do this with particular focus on the basic visualization algorithms.

Date: Wednesday, April 27, 2005
Time: 10:00 AM
Place: 646-PGH

Faculty, students, and the general public are invited.
Thesis Advisor: Dr. Olin Johnson