Kitware Unleashes Brand-New Rendering Backend in ParaView 5.0

Release marks a major milestone in interactive data exploration.

On behalf of the ParaView community, Kitware announces the release of ParaView 5.0. The release follows closely on the heels of ParaView 4.4, building on its features while presenting a rewrite of ParaView’s rendering backend to leverage newer OpenGL features. This major milestone in ParaView’s life cycle ensures that the open-source, multi-platform data analysis and visualization application gets the most out of modern graphics cards and rendering systems for large data sets.

“With each major release of ParaView, we see substantial change,” Utkarsh Ayachit, the lead developer of ParaView at Kitware, said. “In version 3.0, the user interface (UI) switched to Qt; version 4.0 brought the unified Properties panel and modularization; and now, version 5.0 introduces a brand-new rendering backend, marking the first big change to the rendering infrastructure in ParaView’s lifetime.”

ParaView’s new backend uses the Visualization Toolkit’s (VTK’s) redesigned rendering engine. VTK’s engine redesign launched in 2014 with the aim of adopting modern OpenGL application programming interfaces (APIs) and design patterns. Targeted for modern graphics processing units (GPUs), the new engine maintains key capabilities of the old engine, while it significantly improves polygonal and volumetric rendering.

In other rendering news, ParaView 5.0 allows systems with Embedded-System Graphics Library (EGL) support, including those with NVIDIA GPUs and newer drivers, to build ParaView to use hardware-accelerated off-screen rendering without a windowing system (e.g., an X server). For systems without GPUs, ParaView 5.0 binaries for Linux include the initial public release (alpha) of Intel’s Mesa and OpenSWR integration effort. OpenSWR is a software rasterizer that Intel developed for use of unmodified visualization software. OpenSWR runs on laptops, workstations, and compute nodes in high-performance computing (HPC) systems.

ParaView 5.0 also features a pvOSPray plug-in. The plug-in’s Rendered 3D View offers additional display and view properties that provide parameter controls for shadows and material properties, among others. OSPRay is an open-source ray-tracing engine that Intel developed and is specifically optimized for Intel processors. To learn more about efforts to integrate OSPRay and OpenSWR into ParaView, please visit Kitware’s news site.

Another new feature in ParaView 5.0 is a Community Atmosphere Model version 5 (CAM5) ParaView Catalyst adaptor, which enables in situ analysis for CAM5 simulations. ParaView Catalyst is a lightweight version of the ParaView server for HPC co-processing.

Furthermore, ParaView 5.0 unveils the AcceleratedAlgorithms plug-in. Made available for early testers, the plug-in is highly scalable for large data. Two algorithms included in the plug-in, FlyingEdges2D and FlyingEdges3D, generate isosurfaces from two-dimensional (2D) and three-dimensional (3D) data, respectively. The plug-in also contains the FlyingEdgesPlaneCutter algorithm for cutting volumes with a single plane.

“With this release, we have laid the groundwork for modernizing the rendering engine and have kept an eye toward the future of HPC and visualization,” Ayachit said.

Created to analyze extremely large data sets using distributed memory computing resources, ParaView presents developers with the flexibility to tailor functionality to specific problem domains. Today, ParaView is an integral tool in many national laboratories, universities, and industries. The application recently added another award to its accolades with an honor by HPCwire for Readers’ Choice – Best HPC Visualization Product or Technology.

For information on leveraging ParaView and its capabilities, and to download version 5.0, please visit ParaView’s website.

Portions of this work were developed in collaboration by Intel® Corporation, Kitware, and the Texas Advanced Computing Center (TACC) through the Intel Parallel Computing Centers code modernization program.

Research reported in this publication was supported by the National Institute Of Biomedical Imaging And Bioengineering of the National Institutes of Health under Award Number R01EB014955. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

This material is based upon work supported by the U.S. Department of Energy, under Award Number DE-SC0007440 and the U.S. Department of Energy, Office of Science, under Award Number DE-SC0012387.

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.


Questions or comments are always welcome!