Trame: A Python Framework for Interactive Scientific Visualization and Analysis
Trame is an open-source Python framework developed by Kitware for building interactive web-based visualization and analysis applications with minimal frontend expertise. By enabling complete application design—including 2D/3D rendering, widget layout, and event handling—entirely in Python, trame eliminates the complexity traditionally associated with JavaScript-based web development. It integrates tightly with leading visualization platforms such as ParaView and VTK, and supports both client-side and remote (server-side) rendering, making it suitable for High-Performance Computing (HPC), cloud, Jupyter Notebooks, and workstation environments. Trame is also deeply compatible with modern Python-based scientific software, integrating seamlessly with tools such as MOOSE for multiphysics simulation, PyTorch for artificial intelligence (AI), and computational fluid dynamics (CFD) simulators. With real-time synchronization via WebSockets and a reactive state engine, trame empowers developers to build responsive UIs for diverse scientific domains. Example applications include Peacock, a graphical interface for configuring, executing, and analyzing MOOSE simulations; Trame XAITK, for visualizing saliency map to explain AI model inference; and ArrowFlow, for CFD micro-workflow simulations built on M-Star CFD™. By lowering the development barriers to interactive scientific visualization and integrating with the broader scientific software stack, trame accelerates insight, reproducibility, and innovation in computational science and engineering.
IMPACT
Accomplishment: Delivered a Python-only programming framework for integrating modeling, simulation, analysis, and AI in web-based workflow applications; Result: Produced three integrated applications that are used for non-experts to accelerate decision making in engineering and operations.
PRESENTER
Quammen, Cory
cory.quammen@kitware.com
919-308-1306
Kitware, Inc.
CO-AUTHOR(S)
Jourdain, Sebastien
sebastien.jourdain@kitware.com
O'Leary, Patrick
patrick.oleary@kitware.com
Quammen, Cory
cory.quammen@kitware.com
CATEGORY
Visualization
SECONDARY CATEGORY
Mod, Sim & Analysis for Decision Making
SYSTEM(S) USED
None, but this work would be able to run on any HPCMP system