Constantine Pavlakos

Sort-last parallel rendering for viewing extremely large data sets on tile displays

Due to the impressive price-performance of todays PC-based graphics accelerator cards, Sandia National Laboratories is attempting to use PC clusters to render extremely large data sets in interactive applications. This paper describes a sort-last parallel rendering system running on a PC cluster that is capable of rendering enormous amounts of geometry onto high-resolution tile displays by taking advantage of the spatial coherency that is inherent in our data. Furthermore, it is capable of scaling to larger sized input data or higher resolution displays by increasing the size of the cluster. Our prototype is now capable of rendering 120 million triangles per second on a 12 mega-pixel display.

PVR: high-performance volume rendering

Traditional volume rendering methods are too slow to provide interactive visualization, especially for large 3D data sets. The PVR (parallel volume rendering) system implements parallel volume rendering techniques that speed up the visualization process. Moreover, it helps computational scientists, engineers, and physicians to more effectively apply volume rendering to visualization tasks. The authors describe the PVR system that they have developed in a collaboration between the State University of New York at Stony Brook and Sandia National Laboratories. PVR is an attempt to provide an easy-to-use portable system for high performance visualization with the speed required for interactivity and steering. The current version of PVR consists of about 25000 lines of C and Tcl/Tk code. It has been used at Stony Brook, Sandia, and Brookhaven National Labs to visualize large data sets for over a year.

Three dimensional visualization of proteins in cellular interactions

In researching the communication mechanisms between cells of the immune system, visualization of proteins in three dimensions can be used to determine which proteins are capable of interacting with one another at a given time by showing their spatial colocality. Volume data sets are created using digital confocal immunofluorescence microscopy. A variety of visualization approaches are then used to examine the interactions. These include volume rendering, isosurface extraction, and virtual reality. Based on our experiences, we have concluded that no single one of these approaches provides a complete solution for visualizing biological data. However, in combination, their respective strengths complement one another to provide an understanding of the data.

An architecture and implementation to support large-scale data access in scientific simulation environments

At Sandia National Laboratories, a Data Services system has been developed to provide web-based access to high-performance computing clusters that host a set of post-processing applications for very large-scale data manipulation and visualization. A three-tier architecture provides a meta-framework for a collection of smaller frameworks, each of which satisfies a particular aspect of the overall system, including frameworks for a common data model, distributed resource management, component-based software on the cluster, and security. A prototype implementation has been completed which demonstrates the use of all of these frameworks in an integrated environment to provide end users the ability to manage and understand simulation results for very large, complex problems.

A visualization model for supercomputing environments

A description is given of a visualization environment that integrates a supercomputer, large storage facilities, and a visualisation server to let scientists analyze large, complex problems and view pseudo-high-performance, interactive graphics on desktop displays. The paper covers the shared, centralized server, the current prototype environment, and the software environment.<<ETX>>

25 – Desktop Delivery: Access to Large Datasets

This chapter defines large data as datasets that are much greater than the memory capacity of the desktop machine. Transferring very large datasets can require hours or days, and the reality of network errors and machine reboots can turn a planned 8-hour project into a 5-day marathon. The desktop-delivery problem is one of engineering a solution to provide meaningful interaction with a large dataset using a relatively small computer connected by a relatively small network. This chapter presents an approach for delivering supercomputer results to offices. The approach was developed to visualize simulation results produced by the U.S. Department of Energys Accelerated Strategic Computing Initiative. The system design presented in the chapter is the instantiation at Sandia National Labs; similar systems have been implemented at Los Alamos and Lawrence Livermore national labs. Most of the discussion in the chapter presents methods for generating and delivering images to leverage the human visual system for analysis.

28 – Issues and Architectures in Large-Scale Data Visualization

This chapter takes a high-level, end-to-end, systems view of the large-scale data visualization problem in high-performance computing environments. The chapter introduces many of the issues associated with the problem as well as solution approaches and architectural features that are either proven or have promise. Successes in the advanced simulation and computing visual interactive environment for weapons simulation program are used to help provide credence to some of the architectural elements. The presents some of the simple statistics relating to the migration of visualization data. Statistics such as these are particularly relevant in the consideration of remote data analysis and visualization of large data, or visualization in high-performance computing environments that are distributed. The chapter also presents a functional system architecture for high-performance computing environments that can be used to provide the services and end-to-end system characteristics. The architecture links the key components of computational resources, data archival, large high-performance online storage, data service resources, visualization service resources, and the end users display and/or workstation interface.

A visualization environment for supercomputing-based applications in computational mechanics

The authors characterize a visualization environment that has been designed and prototyped for a large community of scientists and engineers, with an emphasis in supercomputing-based computational mechanics. The proposed environment makes use of a visualization server concept to provide effective, interactive visualization to the users desktop. Benefits of using the visualization server approach are discussed. Some thoughts regarding desirable features for visualization server hardware architectures are also addressed. A brief discussion of the software environment is included. The authors summarize certain observations which they have made regarding the implementation of such visualization environments.