Load balancing and computing strategies in pipeline optimization for parallel visualization of 3D irregular meshes

Authors:
Andrea Clematis;Daniele D’Agostino;Vittoria Gianuzzi
Affiliations:
IMATI-CNR, Genova, Italy;IMATI-CNR, Genova, Italy;DISI University of Genova, Genova, Italy
Venue:
PVM/MPI'05 Proceedings of the 12th European PVM/MPI users' group conference on Recent Advances in Parallel Virtual Machine and Message Passing Interface
Year:
2005

Citing 5
Cited 1

Marching cubes: A high resolution 3D surface construction algorithm

SIGGRAPH '87 Proceedings of the 14th annual conference on Computer graphics and interactive techniques
Surface simplification using quadric error metrics

Proceedings of the 24th annual conference on Computer graphics and interactive techniques
The Grid and Future Visualization System Architectures

IEEE Computer Graphics and Applications
Grid-Distributed Visualizations Using Connectionless Protocols

IEEE Computer Graphics and Applications
A Web-Based Isosurface Extraction System for Heterogeneous Clients

EUROMICRO '03 Proceedings of the 29th Conference on EUROMICRO

A parallel protein surface reconstruction system

International Journal of Bioinformatics Research and Applications

Quantified Score

Hi-index	0.00

Visualization

Abstract

Parallel visualization is assuming an increasing role in the deployment of Web and Grid based systems for scientific applications. The visualization process consists of a set of filters or components that are executed in a pipelined assembly that should be adaptively configured on the basis of user requirements, processing architecture and network characteristics. In this paper we focus our attention on the visualization of 3D irregular meshes produced by the interrogation of volumetric data using an isosurface extraction algorithm. We consider a simplified pipeline consisting of two components: isosurface extraction, and mesh simplification. We show that also in this simple case an in-deep analysis is necessary in order to optimize the whole pipeline. In fact different implementation and load balancing strategies are possible for each single component, but the whole pipeline optimization could be achieved combining non-optimal implementation of individual stages. Moreover the quality of the produced mesh should be considered in the selection of an adequate component implementation. The proposed analysis permits to point out trade-offs and algorithmic requirements that should be considered in the design of a complete visualization system for advanced Grid applications.