Scientific visualization with ParaView

July 17^th, 9:00am

Abstract: ParaView is an open source, multi-platform data analysis and visualization tool designed to run on a variety of hardware from individual laptops to large supercomputers. With ParaView, users can interactively visualize 2D and 3D datasets defined on structured, adaptive and unstructured meshes or particles, animate these datasets over time, and manipulate them with a variety of filters. ParaView supports both interactive visualization through a graphical user interface (GUI) and scripted visualization, including offscreen rendering, and is an easy and enjoyable tool to learn.

In this beginner-friendly workshop, you will learn the fundamentals of ParaView, starting with its interface and the types of data it can import. We will build increasingly sophisticated visualization pipelines using filters, then explore ParaView’s Python capabilities by converting GUI actions into scripts, modifying them, and running them. Next, we will create keyframe animations. Finally, we will cover remote and large-scale visualization on HPC clusters for situations where downloading data to a local computer is impractical or impossible. Along the way, we will discuss strategies for optimizing datasets for parallel I/O and the compute resources needed to process very large datasets.

Instructor: Alex Razoumov (SFU)

Workshop materials

The slides for this workshop are included into the main ZIP file (~30 MB), along with sample datasets and various scripts. The slide deck has two parts: introductory slides1.pdf and more advanced slides2.pdf.

Prerequisites

Please install ParaView 6.0.x from https://www.paraview.org/download on your computer before the workshop. No prior ParaView experience required.

Note

Although ParaView 6.1 is available, there are some issues with it at least in MacOS, so we’ll use 6.0 today.

Other materials

• In winter 2026, we ran 18 visualization workshops under the Winter Visualization Series, covering Matplotlib, Seaborn, Plotly (including Dash), Vega-Altair, the Grammar of Graphics in both R and Python, all ParaView topics, VMD, network visualization with Gephi, in-situ visualization with Catalyst2, and modern web-based visualization using trame.
• We recorded ~30 visualization webinars as part of the SFU Research Computing Group’s training program.
• AR leads the Digital Research Alliance of Canada National Visualization Team: check out the gallery and past visualization contests.

Videos

In 2020, we recorded our full-day ParaView workshop and split it into short video clips, which are available below. If anything is unclear after today’s workshop or you would like to revisit the material afterwards, these videos are still an excellent resource. Please keep in mind, however, that some of this content may now be outdated.

Intro

• Basic concepts (7 min)
• 1D/2D plotting (8 min)
• Multi-dimensional sci-vis (11 min)

File formats and workflows

• Running ParaView (11 min)
• File formats and reading raw binary (7 min)
• VTK file formats: overview and legacy VTK (13 min)
• VTK file formats: XML VTK (4 min)
• Scientific file formats (5 min)
• ParaView filters (11 min)
• Store your visualization workflow with a state file (2 min)
• Side-by-side visualization (3 min)
• Visualizing vectors (6 min)
• Creating better streamlines (3 min)
• Line integral convolution (LIC) (3 min)
• Reading CSV data (6 min)
• Putting your visualization online with ParaView Glance (5 min)

Animation and scripting

• Keyframe animation (11 min)
• Other ways to produce animations (3 min)
• Intro to ParaView scripting (16 min)
• Camera animation with scripting (10 min)
• Programmable filter (4 min)

Remote visualization

• Intro to remote visualization (11 min)
• Small-scale remote visualization (10 min)
• Large-scale remote visualization (11 min)
• Using cluster GPUs for rendering (5 min)