The American Museum of Natural History in New York City has built a new exhibit space—The Hall of Planet Earth. This hall highlights earth processes using various exhibits including actual rocks and core samples, demonstration models, and video display stations. One specific scientific area that the museum wants to highlight is that of modeling and simulation. Los Alamos has a long history in this area through our involvement in programs such as the DOE Grand Challenges and the Institute for Geophysics and Planetary Physics. Because of this, we were asked to participate in the design of, and provide content for, five exhibits designed to showcase modeling and simulation of individual earth processes. This paper briefly describes the scientific visualizations developed and used to model atmosphere, ocean, and mantle processes for the American Museum of Natural History’s exhibit.

State of the art scientific simulations are currently working with data set sizes on the order of a billion cells. Parallel rendering is a promising approach for interactivel y isualizing multiple isosurface variables from data sets of this magnitude. In sort-last rendering, each pr o essor creates a depth buffered image of its assigned objects. All processors’ images are composited toget her to create a final result. Improving the efficiency of this compositing step is key to interactiv e parallel rendering. This paper presents a compression-based image compositing algorithm which can prov ide significant savings in both communication and compositing costs.

Sphere rendering is an important method for visualizing molecular dynamics data. This paper presents a parallel algorithm that is almost 90 times faster than current graphics workstations. To render extremely large data sets and large images, the algorithm uses the MIMD features of the supercomputers to divide up the data, render independent partial images, and then finally composite the multiple partial images using an optimal method. The algorithm and performance results are presented for the CM-5 and the T3D.

The recent years have seen rapid advancement towards viewing the Earth as an integrated system. This means that we have come to understand the interdependence of the major planetary subsystems-atmosphere, biosphere, oceans and the deep earth interior-on a large range of time and length scales. One of the longest time scales of the planet is imposed by solid state convection within the silicate Earth mantle. Mantle convection modeling, and other earth science modeling efforts, now are producing simulation data on grids that are large enough to strain the memory and processing power of even the largest high-end graphics workstations. Another alternative is to use parallel visualization tools running on the massively parallel computers that generated the data. This is the approach that we have taken for the visualization of mantle convection simulation data.

Coherency is one of the most important factors that influences the performance of distributed ray tracing systems, especially when object dataflow approach is employed. Designing algorithms that exploit ray-to-ray, spatial, and temporal coherency result in highly scalable systems, as demonstrated by the two algorithms described in this paper. The coherent nature of our algorithms are also delineated by the excellent speedups achieved. Future efforts in parallel rendering should focus on incorporating other types of coherency as well, in our quest to attain the final goal of perfectly linear scalability. 8 compressed cells, 35000 cells will undergo compression and decompression in the case of 1 frame/phase as opposed to 4000 cells in the case of 30 frames/phase. The efficiency and speedup graphs of the ray-front algorithm for different volumes are shown in Figure 8. The algorithm demonstrates about 80% efficiency for 32 processors for all the test volumes. The timings for the 256 3 volumes could not be taken on a single processor due to lack of memory to fit the complete database. The speedups and efficiency are thus measured relative to that of 2 processors. The good speedup also suggests that considerable load-balancing has been achieved using the static block-cyclic scheme. Although the algorithm demonstrate considerable speedup, there is scope to further improve its performance. Currently, the FTBL is determined in a rather conservative manner, and a number of cells are fetched irrespective of whether they are needed in the future or not. If a more accurate determination of these cells is possible, then some cell requests and the associated overheads can be avoided. We are currently trying to accomplish a more efficient scheme for improved latency hiding. Dynamic load balancing for the rayfront algorithm is another area of future research. At the time of running these experiments, all 128 processors on the T3D were not available to a single user. We will report these results in the final paper. Object dataflow approaches are known to take advantage of object-space coherency to reduce the number of misses. For sufficient local memory sizes, this helps to reduce the amount of communication between processors compared with ray-dataflow approaches. The hit ratio can further be increased if frame-to-frame coherence is also exploited. The hit ratio directly effects the overheads incurred by the system. Each object fetched from non-local memory incurs a start-up cost, and if it is not a pre-fetch data acquisition scheme, …

This paper presents a data distributed parallel ray-traced volume rendering algorithm and its implementation on the CRI T3D. This algorithm distributes the data and the computational load to individual processing units to achieve fast and high-quality rendering of high-resolution data. The volume data, once distributed, is left intact. The processing nodes perform local raytracing of their subvolume concurrently. No communication between processing units is needed during this local ray-tracing process. A subimage is generated by each processing unit and the nal image is obtained by compositing subimages in the proper order by the Binary-Swap algorithm. Performance of this algorithm on the T3D is presented and compared to an implementation on the CM-5.

This paper presents the Advanced Computing Lab Message Passing Library (ACLMPL). Modeled after Thinking Machines Corporation`s CMMD, ACLMPL is a high throughout, low latency communications library for building message passing applications. The library has been implemented on the Cray T3D, Thinking Machines CM-5, SGI workstations, and on top of PVM. On the Cray T3D, benchmarks show ACLMPL to be 4 to 7 times faster than MPI or PVM.

We describe a parallel volume-rendering algorithm, which consists of two parts: parallel ray tracing and parallel compositing. In the most recent implementation on Connection Machine's CM-5 and networked workstations, the parallel volume renderer evenly distributes data to the computing resources available. Without the need to communicate with other processing units, each subvolume is ray traced locally and generates a partial image. The parallel compositing process then merges all resulting partial images in depth order to produce the complete image. The compositing algorithm is particularly effective for massively parallel processing, as it always uses all processing units by repeatedly subdividing the partial images and distributing them to the appropriate processing units. Test results on both the CM-5 and the workstations are promising. They do, however, expose different performance issues for each platform.

-This paper discusses the use of general-purpose graphics workstations for interactive high-resolution volume visualization. We survey previous research results in parallel volume rendering as well as commercial products that take advantage of parallel processing to make volume rendering a practical visualization method. Our focus is on developing distributed computation methods that can distribute the memory and computational demands of volume visualization across a network of general purpose workstations. We describe three distributed computation strategies based on ray-casting volume rendering that can be implemented on either shared-memory multiprocessor workstations or on a network of ordinary workstations. Multiple views of real-time feature extraction give tremendous insight to the volume data. Multiple variable visualization helps scientists to capture the interaction between important variables in a simulation. Divideand-conquer rendering allows interactive high-resolution volume visualization of large data sets on a network of midrange workstations, even when the data set is too large for available memory on any single workstation. Several examples in medical imaging and computational fluid dynamics are shown illustrating the practicality of these methods. 1. INTRODUC'I'ION The advance of parallel architecture and software carties computational science into a new dimension in which scientists have been able to more accurately approximate and explore the world around us and thus derive many new scientific discoveries. Scientific visualization, the use of computer graphics to provide visual interpretations of simulations of physical phenomena or acquired data from scanning systems, often requires intensive computation and the management of enormous amount of data. Over the past few years, highly commercial and academic interests have led to the development of many new parallel architectures and software algorithms for data visualization [ l, 2, 3 ]. Vector supercomputers like the Cray and distributed-memory supercomputers like the Connection Machine are expensive to maintain and run, and usually must be shared between many users. Other specialpurpose architectures, developed in research environments, are also not generally available to scientists who may need to conduct visualization procedures daily in their laboratory. The increase in performance and decrease in price of general-purpose graphics workstation suggest that we could make good use of them for not only numerical simulations but also data visualization. In this paper, we describe how general-purpose graphics workstations, particularly multiprocessor and networked workstations, can be used to visualize large or multiple data sets, and to do interactive feature extraction and viewing. We show examples in medical imaging and combustion simulations making use of IRIS 4D/240 GTX or multiple IBM RS/6000 Model 520/530 workstations to make the process of data visualization much more effective and efficient. The examples shown here use IVES (Interactive Volume Exploration System) [4, 5 ], an visualization system based on ray-casting volume rendering that we have developed for local scientists. 2. PARALLEL VOLUME VISUALIZATION Most scientific and biomedical data sets are scalar or vector fields of three spatial dimensions known as volume data. Direct volume rendering, creating an image directly from volume data without constructing intermediate graphics primitives, has been shown to be a very effective method for visualizing scalar volume data[6, 7, 8]. However, volume rendering is computational expensive and the rendering time grows linearly with the size of the data set. As a result, many algorithms have developed which take advantage of the coherence in the data and the rendering process [ 4, 9 ]. Parallel volume rendering schemes have also been developed, using special-purpose hardware [ l 0, 11 ] or supercomputers [ 12, 13 ]. The major algorithmic strategy for parallelizing volume rendering is the divide-and-conquer paradigm. The volume rendering problem can be divided either by data space subdivision (DSS) or by image space subdivision (ISS). While DSS assigns the computation associated with particular subvolumes to processors, ISS distributes the computation associated with particular portions of the image space. DSS is usually implemented on a distributed-memory parallel computing environment. On the other hand, ISS is most efficient on a shared-memory multiprocessor computer. Hybrid methods are also feasible. Splatting is a data-space parallel volume rendering algorithm that maps the volume data onto the image plane. A full resolution rendering of a data set with 96 X 128 X 113 grid points takes about one minute on either a Sun TACC-1 or four Sun-4s[ 14]. As splatting is better suited for coarse to medium grain parallelism, a multi-pass shear decomposition algorithm has been implemented on the finely distributed-memory Connection Machine [ 12 ] to approximate real-time rotation of the volume data. The rendering uses parallel computation constructs offered by the Connection Machine

This paper discusses the use of general-purpose graphics workstations for interactive high-resolution volume visualization. We survey previous research results in parallel volume rendering as well as commercial products that take advantage of parallel processing to make volume rendering a practical visualization method. Our focus is on developing distributed computation methods that can distribute the memory and computational demands of volume visualization across a network of general purpose workstations. We describe three distributed computation strategies based on ray-casting volume rendering that can be implemented on either shared-memory multiprocessor workstations or on a network of ordinary workstations. Multiple views of real-time feature extraction give tremendous insight to the volume data. Multiple variable visualization helps scientists to capture the interaction between important variables in a simulation. Divide-and-conquer rendering allows interactive high-resolution volume visualization of large data sets on a network of midrange workstations, even when the data set is too large for available memory on any single workstation. Several examples in medical imaging and computational fluid dynamics are shown illustrating the practicality of these methods.

Article Free Access Share on Radioptimization: goal based rendering Authors: John K. Kawai View Profile , James S. Painter View Profile , Michael F. Cohen View Profile Authors Info & Claims SIGGRAPH '93: Proceedings of the 20th annual conference on Computer graphics and interactive techniquesSeptember 1993 Pages 147–154https://doi.org/10.1145/166117.166136Online:01 September 1993Publication History 73citation584DownloadsMetricsTotal Citations73Total Downloads584Last 12 Months12Last 6 weeks1 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteeReaderPDF

This paper presents a divide-and-conquer ray-traced volume rendering algorithm and a parallel image compositing method, along with their implementation and performance on the Connection Machine CM-5, and networked workstations. This algorithm distributes both the data and the computations to individual processing units to achieve fast, high-quality rendering of high-resolution data. The volume data, once distributed, is left intact. The processing nodes perform local raytracing of their subvolume concurrently. No communication between processing units is needed during this locally ray-tracing process. A subimage is generated by each processing unit and the final image is obtained by compositing subimages in the proper order, which can be determined a priori. Test results on the CM-5 and a group of networked workstations demonstrate the practicality of our rendering algorithm and compositing method.

8.1. IntroductionThe problem of passing a surface through a set of data points arises in numerous areas of application such as medical imaging, geological modeling, scientific visualization, and geometric modeling. Variants of this problem have been approached from many directions. Tensor-product B-splines work well for modeling surfaces based on rectilinear control nets but are not sufficient for more general topological types. Triangulated data, however, can represent arbitrary topological types. In this paper, we present a survey of a class of schemes that address the problem of fitting a surface to triangulated data.

Article Volume seedlings Share on Authors: Michael F. Cohen View Profile , James Painter View Profile , Mihir Mehta View Profile , Kwan-Liu Ma View Profile Authors Info & Claims I3D '92: Proceedings of the 1992 symposium on Interactive 3D graphicsJune 1992 Pages 139–145https://doi.org/10.1145/147156.147184Online:01 June 1992Publication History 20citation376DownloadsMetricsTotal Citations20Total Downloads376Last 12 Months3Last 6 weeks0 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteeReaderPDF

Image Synthesis is not a new idea. Human have been creating pictures of non-existent worlds for as long as they have had the capacity to imagine them. These efforts have ranged from cave drawings to fantastically rendered scenes of great complexity. They have depicted scenes of great horror in the paintings of Hironymous Bosch, to the impossible worlds of M. C. Escher.

AbstractRay‐traced volume rendering has been shown to be an effective method for visualizing 3D scalar data. However, with currently available workstation technology, interactive volume exploration using conventional volume rendering is still too slow to be attractive. This paper describes an enhanced volume rendering method which allows interactive changes of rendering parameters such as colour and opacity maps. An innovative technique is provided which allows the user to plant a ‘seed’ in the volume to rapidly modify local shading parameters. For a fixed viewing position, the user can interactively explore specific regions of interest. Furthermore, a virtual cutting technique with the exploratory seed allows the user to remove surfaces and see the internal structure of the volume. Examples demonstrate these techniques as an attractive option in many applications.