Advanced visualization collaboration resources

Many of the following resources are PDF files. To obtain a free copy of the latest Adobe Acrobat Reader, go to the Adobe download page.

- Documentation - Links - Presentations -
- Press releases - Technical reports - White papers

Supporting Multi-CPU and GPU Systems in AVS/Express
[2.43Mb .PPT]

AVS/Express is a long established, leading visualization development application however it was developed for the single processor workstation. Over the years the University of Manchester has been involved in developing support in AVS/Express for CAVE and Reality Centre environments and more recently parallel computation systems. The presentation will give an overview of development and focus on more recent work; the Distributed Data Renderer Edition which uses the Parallel Compositing Library.

Source: James Perrin, University of Manchester.

Simulation and Rendering of Fluids on a GPU Cluster

This presentation reviews the solution alternatives of the Navier-Stokes equations and presents real-time fluid simulation and visualization approaches exploiting the computational power of a GPU cluster to display fluids interactively. Both Eulerian and Lagrangian techniques are covered and compared. We use object space decomposition of the 3D volume. Simulation data are exchanged by MPI and the framelets generated by the nodes are composited by the Parallel Compositing library. The following link provides the full abstract, presentation, videos, and source for the programs.
Source: László Szirmay-Kalos, TU Budapest

» HP CCN Call

Compositing Using COTS Components [190 Kb .PDF]

This presentation describes HP's vision for image compositing using strictly commercial, off-the-shelf (COTS) components. HP recently shared this vision through this HP CCN forum. The HP Viz team is seeking confirmation from users and ISVs that this approach is in alignment with your needs. Please provide us with your feedback.
Source: HP Visualization Team

Stony Brook Visual Computing Cluster [5.2Mb .PPT]

Presentation outlining the objectives, specifications, and results of the Visual Computing Cluster at Stony Brook.
Source: Susan Frank, Stony Brook Center for Visual Computing

Parallelldatorcentrum, PDC [1.37 MB PPT]

Summary of ongoing simulation and visualization projects at national academic supercomputer center at KTH
Source: PDC of the Swedish Royal Institute of Technology (KTH)

High End Visualization for Performance Computing (PDF)

This presentation describes the rationale, vision, markets, and technology of the Sepia Visualization System.
Source: High Performance Technical Computing Division of Hewlett-Packard Corporation

This document describes the current version of the Sepia Compositing Network as implemented for Red Hat Linux V7.3, including a coding example and the API call formats. It represents “work in progress” and is in pre-product form. It does not represent any commitment by HP to release a product with these features.

» http://www.llnl.gov/asci/views/views.html
Link to Lawrence Livermore's ASCI VIEWS program, including details on the technology and goals of the program.

» SHARCNET helping researchers 'beat the clock'

Canada's largest High Performance Computing (HPC) consortium recently launched its second phase of development at concurrent events at Western, McMaster University and the University of Guelph.

HP brochure: "HP is developing technologies for enterprise Grids that support mission-critical applications."

» http://www.hp.com/hpinfo/newsroom/press/2002/021028a.html
Press release on the award by the Dept. of Energy to HP to develop the visualization system required by ASCI VIEWS.

by Susan Frank and Arie Kaufman, Center For Visual Computing and Department of Computer Science, Stony Brook University

This paper describes rendering of massive data sets on a volume visualizaton cluster and the technique developed for reducing and managing volume data.

The Parallel Compositing Library

The Parallel Compositing Library: This white paper looks at distributing the generation of graphical images over a set of computers. The paper describes the Parallel Compositing Library defined and implemented by HP in collaboration with several partners. The intent of this Library is to provide a common set of functions that facilitate parallel rendering.

SEPIA - Applicability to MVC ( PDF - 1.2 MB)

by Paul G. Lever, Project Manager, Manchester Visualization Centre (MVC), University of Manchester

This paper provides a brief overview of the SEPIA technology from HP, detailing the hardware components, software support, and overall aim of the system and how this compares with the currently installed systems.The flexibility offered by SEPIA is discussed and how this may provide a better delivery system
for use by staff and users of the Visualization Immersion Projection Laboratory (VIPL).

Distributed rendering of interactive soft shadows ( PDF - 247 KB)

by M. Isardy, M. Shand, and A. Heirich (Compaq Computer Corporation)

Editor's note: The following three papers describe the rationale and evolution of the technology that underpins the Sepia Visualization System under development by the HPTC Division of HP.

Scalable Interactive Ray-Casting of Volumes Using Off-the-Shelf Components ( PDF ) ( HTML )

by Santiago Lombeyda, Mark Shand, Laurent Moll, David Breen, and Alan Heirich

This paper reports on the application of a second generation implementation (Sepia-2) of this architecture to structured volume visualization. The goal of this current work is to demonstrate that large volume datasets (3D rectilinear grids of scalars) can be interactively rendered with a networked configuration of off-the-shelf components.

Scalable Distributed Visualization Using Off-the-Shelf Components( PDF ) ( HTML )

by Alan Heirich, Laurent Moll

This paper describes a visualization architecture for scalable computer systems. The architecture is currently being prototyped for use in Beowulf-class clustered systems.

Sepia: scalable 3D compositing using PCI Pamette ( PDF ) ( HTML )

by Alan Heirich, Laurent Moll, Mark Shand

Describes an implementation of an image combining architecture that allows distributed rendering of a partitioned data set at interactive rates. The architecture achieves real-time frame rates and low latency through pipelining and the use of a high bandwidth network technology to transfer the image data. It is flexible because it uses programmable FPGA devices to implement the combining logic.