We present a new technique for providing interpolation within cell-centered Adaptive Mesh Refinement (AMR) data that achieves C(0) continuity throughout the 3D domain. Our technique improves on earlier work in that it does not require that adjacent patches differ by at most one refinement level. Our approach takes the dual of each mesh patch and generates "stitching cells" on the fly to fill the gaps between dual meshes. We demonstrate applications of our technique with data from Enzo, an AMR cosmological structure formation simulation code. We show ray-cast visualizations that include contributions from particle data (dark matter and stars, also output by Enzo) and gridded hydrodynamic data. We also show results from isosurface studies, including surfaces in regions where adjacent patches differ by more than one refinement level.

We describe a concurrent visualization pipeline designed for operation in a production supercomputing environment. The facility was initially developed on the NASA Ames "Columbia" supercomputer for a massively parallel forecast model (GEOS4). During the 2005 Atlantic hurricane season, GEOS4 was run 4 times a day under tight time constraints so that its output could be included in an ensemble prediction that was made available to forecasters at the National Hurricane Center. Given this time-critical context, we designed a configurable concurrent pipeline to visualize multiple global fields without significantly affecting the runtime model performance or reliability. We use MPEG compression of the accruing images to facilitate live low-bandwidth distribution of multiple visualization streams to remote sites. We also describe the use of our concurrent visualization framework with a global ocean circulation model, which provides a 864-fold increase in the temporal resolution of practically achievable animations. In both the atmospheric and oceanic circulation models, the application scientists gained new insights into their model dynamics, due to the high temporal resolution animations attainable.

This work describes the methods used to produce an interactive visualization of a 2 TB computational fluid dynamics (CFD) data set using particle tracing (streaklines). We use the method introduced by Bruckschen el al. (2001) that precomputes a large number of particles, stores them on disk using a space-filling curve ordering that minimizes seeks, then retrieves and displays the particles according to the user's command. We describe how the particle computation can be performed using a PC cluster, how the algorithm can be adapted to work with a multiblock curvilinear mesh, how scalars can be extracted and used to color the particles, and how the out-of-core visualization can be scaled to 293 billion particles while still achieving interactive performance on PC hardware. Compared to the earlier work, our data set size and total number of particles are an order of magnitude larger. We also describe a new compression technique that losslessly reduces the amount of particle storage by 41% and speeds the particle retrieval by about 20%.

We present techniques for discovering and exploiting regularity in large curvilinear data sets. The data can be based on a single mesh or a mesh composed of multiple submeshes (also known as zones). Multi-zone data are typical in Computational Fluid Dynamics (CFD) simulations. Regularities include axis-aligned rectilinear and cylindrical meshes as well as cases where one zone is equivalent to a rigid body transformation of another. Our algorithms can also discover rigid-body motion of meshes in time-series data. Next, we describe a data model where we can utilize the results from the discovery process in order to accelerate large data visualizations. Where possible, we replace general curvilinear zones with rectilinear or cylindrical zones. In rigid-body motion cases, we replace a time-series of meshes with a transformed mesh object where a reference mesh is dynamically transformed based on a given time value in order to satisfy geometry requests, on demand. The data model enables us to make these substitutions and dynamic transformations transparently with respect to the visualization algorithms. We present results with large data sets where we combine our mesh replacement and transformation techniques with out-of-core paging in order to achieve analysis speedups ranging from 1.5 to 2.

In the area of scientific visualization, input data sets are often very large. In visualization of computational fluid dynamics (CFD) in particular, input data sets today can surpass 100 Gbytes, and are expected to scale with the ability of supercomputers to generate them. Some visualization tools already partition large data sets into segments, and load appropriate segments as they are needed. However, this does not remove the problem for two reasons: 1) there are data sets for which even the individual segments are too large for the largest graphics workstations, 2) many practitioners do not have access to workstations with the memory capacity required to load even a segment, especially since the state-of-the-art visualization tools tend to be developed by researchers with much more powerful machines. When the size of the data that must be accessed is larger than the size of memory, some form of virtual memory is simply required. This may be by segmentation, paging, or by paged segments. The authors demonstrate that complete reliance on operating system virtual memory for out-of-core visualization leads to egregious performance. They then describe a paged segment system that they have implemented, and explore the principles of memory management that can be employed by the application for out-of-core visualization. They show that application control over some of these can significantly improve performance. They show that sparse traversal can be exploited by loading only those data actually required.

We have developed techniques for distributing a hierarchical display list from a PHIGS-like library across a multicomputer. By storing a portion of the database at each processor, inter-processor communication is reduced. This reduction promises traversal of the display list at rates supporting rendering speeds of one million polygons/second or more, as we hope to achieve on our new machine, Pixel-Planes 5 (under construction). Our distribution techniques support order-dependent primitives and allow general display list editing.

This paper introduces the architecture and initial algorithms for Pixel-Planes 5, a heterogeneous multi-computer designed both for high-speed polygon and sphere rendering (1M Phong-shaded triangles/second) and for supporting algorithm and application research in interactive 3D graphics. Techniques are described for volume rendering at multiple frames per second, font generation directly from conic spline descriptions, and rapid calculation of radiosity form-factors. The hardware consists of up to 32 math-oriented processors, up to 16 rendering units, and a conventional 1280 × 1024-pixel frame buffer, interconnected by a 5 gigabit ring network. Each rendering unit consists of a 128 × 128-pixel array of processors-with-memory with parallel quadratic expression evaluation for every pixel. Implemented on 1.6 micron CMOS chips designed to run at 40MHz, this array has 208 bits/pixel on-chip and is connected to a video RAM memory system that provides 4,096 bits of off-chip memory. Rendering units can be independently reasigned to any part of the screen or to non-screen-oriented computation. As of April 1989, both hardware and software are still under construction, with initial system operation scheduled for fall 1989.