NVIDIA IndeX for ParaView Plug-in

                  Interactive Volume Visualization for Faster Discoveries


                  GATHER INSIGHTS FASTER.

                  Visualizing simulation results quickly is essential to scientific discovery. However, existing visualization tools aren’t equipped to deliver a real-time experience on large structured and unstructured datasets.

                  NVIDIA IndeX? is a leading volume visualization tool for HPC that helps to meet this challenge. It takes advantage of the GPU’s computational horsepower to deliver real-time performance on large datasets by distributing visualization workloads across a GPU-accelerated cluster.

                  Now, IndeX is integrated in ParaView—one of the most popular visualization tools in HPC. This means that users have access to IndeX features by default with the latest ParaView version, so they can experience real-time interactivity with their existing volume visualization workflows. And IndeX provides visualization performance that scales with data, so scientists can analyze the entire dataset in its original resolution for faster discoveries.

                  See how NVIDIA IndeX enables real-time transfer function editing in ParaView

                  Watch Now [01:40]

                  FEATURES AND BENEFITS

                  NVIDIA IndeX for ParaView plug-in offers interactive volume visualization that scales with data to enable scientists to analyze the entire dataset for faster time to discovery.

                  Core Collapse Supernova


                  Xtreme Accelerated Computing (XAC) is a feature of the NVIDIA IndeX SDK that allows user-defined compute to be executed at run-time. The ParaView plugin for IndeX takes advantage of the XAC interface and advanced visual presets to gain better insights into the dataset by highlighting key structures. For example, the depth-enhancement preset highlights visual cues based on depth and the edge-enhancement preset highlights silhouettes in the dataset.

                  Dataset courtesy of Dr. Blondin, North Carolina State University.

                  Simultaneous multi-geometry rendering of clouds as volume and wind direction as streamlines

                  Simultaneous multi-geometry rendering of clouds as volume and wind direction as streamlines


                  Data from large-scale simulations typically has information on multiple attributes. Weather simulation, for instance, includes wind speed, cloud data, temperature, and pressure. Each of these attributes are rendered as different primitives to enable better insight. Clouds are rendered as volumes and the wind direction is usually rendered as streamlines or glyphs. The plugin lets users blend ParaView-rendered geometry in a depth-correct way with IndeX volume rendering for better analysis.

                  Dataset courtesy of Niklas R?ber, DKRZ

                  User-defined region isolates the noise revealing horizon structure in the 3D seismic data

                  User-defined region isolates the noise revealing horizon structure in the 3D seismic data

                  USER-DEFINED REGION

                  Visualization is the best way to analyze and extract information from large-scale volumetric data. Even if the entire dataset is interesting, sometimes highlighting key features and visualizing a subset improves productivity. The plugin allows users to freely select the areas of interest, which not only cuts down unwanted information, but also improves performance by only rendering the selected region. For example, in the 3D seismic volumetric dataset shown below, there are noisy regions that obstruct the actual horizon structure in the lower part of the dataset.  Selecting a reduced region of interest helps to bring the horizon in focus for better analysis.

                  Simpler and closer look at volumetric data with 2D slicing

                  Simpler and closer look at volumetric data with 2D slicing


                  Volumetric data holds rich information, but the third dimension also adds complexity to the data analysis. Now, you can simultaneously observe multiple structures in a dataset with multi-planer slicing. Slices of data across multiple parallel and perpendicular planes can be analyzed to identify relational effects, giving a closer look at the key features of the data sets and speeding up analysis. Certain features like vortices in CFD flow simulation use cases that were hard to identify with just volumetric rendering can now be easily analyzed with 2D slices.

                  Dataset courtesy of Jackie Chen, Sandia National Lab.

                  See NVIDIA IndeX deliver real-time interactivity in ParaView [02:02]


                  The IndeX with ParaView plug-in enables scientists to analyze the simulated data at full resolution in real-time by rendering the large-scale data at high frame rates. With the entire data in sight, scientists can quickly converge on the most interesting data for further analysis, resulting in faster time to discovery.

                  See NVIDIA IndeX deliver seamless workflow in ParaView [02:03]

                  SEAMLESS WORKFLOW

                  The IndeX for ParaView plug-in allows users to take advantage of powerful features of IndeX through the ParaView interface. This eliminates the need to learn a new tool, keeps workflow unchanged, and allows users to focus on their research and accelerate discoveries.

                  Data Center IndeX Paraview Plugin GPU Accelerated Performance

                  PERFORMANCE AT SCALE

                  Existing volume visualization tools do not allow scientists to interact with large-scale data in real-time. Scientists either have to wait a few seconds to load each interaction or implement workarounds using iterative process to converge on the most interesting data. IndeX for ParaView plug-in delivers real-time performance, on large datasets by distributing the data across multiple nodes within a GPU-accelerated cluster.


                  OPEN SOURCE PLUG-IN

                  The benefit of open source software is that users can customize the application to fit their needs. Like ParaView, the IndeX plug-in is also open source to easily fit with customized versions of ParaView. The plugin can be compiled as any other standard ParaView plugin without any need for special tools and compilers.

                  Thanks to BSC, we generate large models of human heart. To verify the accuracy of the models, we have to visualize the full dataset and high interactivity visualization is key to gathering meaningful findings. The IndeX plug-in enabled us to visualize our 54M tetrahedral cell model in real-time. Best of all, it fits right inside our existing ParaView workflow.

                  -Mariano Vazquez, Barcelona Supercomputing Center.

                  OTHER FEATURES

                  • Delivers real-time interactivity for large datasets
                  • Depth correct mixing with ParaView primitives
                  • Dynamic streaming and GPU caching of time-varying volume data
                  • Supports all common ParaView data formats
                  • Compositing with ParaView geometries on single node
                  • Workstation edition pre-built for Windows and Linux

                  AND HPC CLUSTERS

                  There are two versions of the plug-in. For usage in a workstation, or single server node, the plug-in is available at no cost. For performance at scale in a GPU-accelerated multi-node system, the Cluster edition of the plug-in is available at no cost to academic users and with a license for commercial users.

                  For more information and to engage with others, visit the IndeX forum page.


                  GET THE CLUSTER EDITION
                  PLUG-IN TODAY. FREE FOR

                  GET FREE WORKSTATION
                  EDITION PLUG-IN