ACADEMIA

Sandia National Labs, Kitware Inc., and NVIDIA Corporation announced a breakthrough in large data scientific visualization, attaining rendering rates of over 1.5 billion polygons per second. The breakthrough was achieved with ParaView (www.paraview.org), an open source visualization application developed by Kitware Inc. (www.kitware.com), processing data in the National Nuclear Security Administration's (NNSA) Advanced Simulation and Computing (ASC) program at the three national laboratories: Sandia National Labs (SNL); Los Alamos National Labs (LANL); Lawrence Livermore National Labs (LLNL). Visualization is an integral component of ASC and is essential to understanding the massive data produced in simulations for national security. One of the world's largest polygonal datasets is a 473 million triangle isosurface generated from a Richtmyer-Meshkov simulation run at LLNL (LLNL: UCRL-MI-151066). In a recent test with this isosurface, Sandia utilized ParaView on 128 visualization nodes (workstations), each comprised of the following: -- NVIDIA Quadro FX 3400 PCI Express graphics boards -- Dell Precision 470 workstations equipped with Dual 3.6 GHz Intel Xeon EM64T processors, and 4GB of RAM -- InfiniBand 4x HCA interconnect, allowing the data to be processed across the 128 nodes ParaView performed various operations on the data including coloring, t-stripping, clipping, and glyphing at interactive rates. Rendering of the surface was performed at an aggregate rate of over 1.5 billion polygons per second, which equates to three-four frames per second. For typical simulation results, ParaView streams images from a cluster to the user's desktop at about 15 frames per second. This level of performance is enabled by the latest generation of NVIDIA graphics hardware and fast PCI Express read-back rates. Sandia currently has over 260 NVIDIA Quadro FX 3400 PCI Express graphics boards installed in its visualization clusters. "The combination of ParaView and high-performance graphics hardware has opened up a new level of interactivity with large data sets, helping researchers around the world to better visualize many types of data -- from global climate modeling to intricate fluid dynamic simulations," said Brian Wylie, visualization team leader, Sandia National Labs. ParaView is also being used by the US Army Research Laboratory (ARL) on tiled display systems for the analysis of physics based simulations in armor/anti-armor applications. ARL's large projection and LCD display walls are driven with NVIDIA Quadro FX 3000G graphics boards installed in AMD Opteron-based visualization clusters. In this case, the US Army is able to leverage both the graphics processing and framelocking capabilities of the NVIDIA hardware to display large datasets across a number of displays or projectors for improved image analysis. "When calculations require tens of CPU years and produce terabytes of output, parallel visualization is no longer a luxury; it's a necessity," said Jerry Clarke, scientific visualization team leader, US Army Research Laboratory. "ParaView on our visualization clusters is an important part or our physics-based simulation environment and our future."