3D Interactive Visualization for Discovery and Exploration
NVIDIA IndeX® is a 3D volumetric visualization SDK that allows scientists and researchers to visualize and interact with massive data sets, make real-time modifications, and navigate to the most pertinent parts of the data, all in real-time, to gather better insights faster. IndeX leverages GPU clusters for scalable, real-time, visualization and computing of multi-valued volumetric data together with embedded geometry data.
- Visualization — Accurate, high-quality data visualization, feature representation and annotation
- Scalability — Interactive 3D visualization of big data and display at any resolution
- Scene management — High-level scene description and management
- Interaction — Infrastructure for the creation of rich visualization interaction techniques
- GPU computing — High-performance volume visualization and analysis using the GPU
- Extensibility — Flexibility and extensibility for a variety of application domains
NVIDIA IndeX 2.1
This new release of NVIDIA IndeX announced and demonstrated at SC’19 is another leap forward for our real-time scalable visualization SDK and includes many exciting new features.
IndeX 2.1 Highlights
- NVIDIA IndeX leverages GPU Direct Storage (GDS), allowing for interactive visualisation of massive datasets (a 150 TB dataset using GDS was demonstrated at SC’19)
- NVIDIA IndeX supports the ARM architecture, which enables HPC visualization on ARM-based GPU clusters
- NVIDIA IndeX is in the IBM Power9 Software Package, ready to be used for visualisation and compute of 3D scientific data
- High-Bandwidth Inference for scientific data through NVIDIA IndeX, IndeX supports interpretation and analysis of large data sets
- NVIDIA IndeX Accelerated Compute enables CUDA-programmable volume shading and computing. The new interface allows interactive modification of the visual appearance and the data itself
- OpenVDB Support with NVIDIA IndeX, is an extension of the data format to enable scientific visualization in the Media and Entertainment domain
- NVIDIA IndeX Finger is a new HTML5 based Viewer, which enables tailor-made cloud and end-user workflows and solutions
- New NVIDIA IndeX Plugin for ParaView 5.7, with improved memory management and data loading time
Contact Us to learn more about this release
Researchers from the UC Santa Cruz and Princeton, use NVIDIA IndeX to visualize a more than 7 TB of an HPC simulation of a dwarf galaxy. With Cholla, the GPU accelerated simulation library, and NVIDIA IndeX, they've discovered how the supernova explosions in the disk of the galaxy couple their energy to the dense gas in the disk of the galaxy, thereby heating it and driving the gas into the dense outflow. Without the visualization capabilities of Index, they couldn't understand the 3D structure of the gas where the explosions interact with the disk. Now they can identify where the action is happening, manipulate the terabytes dataset in real-time, and then animate the results to see how the system evolves. NVIDIA IndeX has been a revolutionary technology for them to run bigger and better simulations.
Researchers from UC Berkeley, the University of Illinois at Urbana, and Caltech, used IndeX to interact and analyze an HPC simulation of a supernova. An HPC simulation of this nature takes about four months and generates over a terabyte of visualization data. Visually analyzing the data is the next big challenge, adding weeks to process and generate batch renderings - making it difficult and adding delays for researchers to gain insights. But with NVIDIA IndeX scientists and researchers can significantly accelerate their visualization pipeline with real-time feedback and interactive data analysis, giving researchers more time to analyze their data and gain better insights faster.
Jet propulsion engineers and scientists used NVIDIA IndeX for visualization, e.g., illustrating the gas mixing out of a methane/oxygen coaxial rocket injector. The flow comprises a core jet of oxygen gas with a ring of methane gas around it. The simulation is based on the following: A Numerical Study of High-Pressure Oxygen/Methane Mixing and Combustion of a Shear Coaxial Injector, Nan Zong & Vigor Yang, AIAA 2005 and the simulation ran on 128 GPU nodes on Titan Supercomputer. The visualization is based on 7.5TB volume dataset running on a GPU cluster (SaturnV predecessor) using NVIDIA IndeX 2016.
Interactive 3D visualization of hundreds of simulated time-steps of an E8 tornado. The visualized database is In total 1+TB. NVIDIA IndeX runs on an NVIDIA VCA cluster and sends a video stream to a client-side viewer.
NVIDIA has partnered with Kitware to bring NVIDIA IndeX to ParaView. Data sets for computational scientists continue to grow from terabytes to petabytes. The IndeX plug-in for ParaView was developed to aid in the analysis and 3D visualizations of extremely large datasets.
NVIDIA GPUs are ideal for simulating large seismic data and ParaView’s comprehensive GUI simplifies multi-platform data analysis and visualization. The HPC community can now visualize high-quality renderings and interactive performance with native visualization functionality such as streamlines, vector plots, etc.
- ParaView plug-in to fully utilize native visualization functionality such as as vector plots, isosurfaces, volume rendering, streamlines, etc.
- GPU accelerated rendering of high-quality and interactive visualizations
- Support for typical grid types, such as structured uniform grids and sparse volumes.
IndeX plug-in for ParaView visualizing database (7+TB) of simulated time-steps of a Galactic Wind.
Contact Us to learn more about this release.