Awesome

Cinema Toolkit, v2.0 (October 2021)

The Cinema toolkit has been developed by Los Alamos National Laboratory, as part of the Cinema Project.

Cinema is an innovative way of capturing, storing and exploring extreme scale scientific data. It is a highly interactive image-based approach to data analysis and exploration that promotes investigation of large scientific datasets, and is easily integrated into your existing workflows through extensions to widely used open source tools. This novel approach supports interactive exploration of a wide range of results, while still significantly reducing data movement and storage.

Extreme scale scientific simulations are leading a charge to exascale computation, and data analytics runs the risk of being a bottleneck to scientific discovery. Due to power and I/O constraints, we expect in situ visualization and analysis will be a critical component of these workflows. Options for extreme scale data analysis are often presented as a stark contrast: write large files to disk for interactive, exploratory analysis, or perform in situ analysis to save detailed data about phenomena that a scientists knows about in advance. With Cinema, we have developed a novel framework for a third option – a highly interactive, compact and scalable way to explore data.

Cinema Data Specification

The Current Data Specification describes the Cinema database data format, which can be easily written and read applications, tools and scripts.

Cinema Toolkit code and repositories

Many Cinema-related repositories can be found on the github cinema group, however, those repositories are not automatically considered to be part of the official supported Cinema releases. Supported Cinema repositories are included here, in the Cinema toolkit.

Update to Cinema 2.0

The Cinema toolkit is updated with this release to track the cinemasci module, where the bulk of new work is now being released. This new model of the Cinema Toolkit depricates command line tools, and embraces a single python module that contains a useful set of components for reading, writing and viewing Cinema databases.

Major updates for this release of the toolkit:

• The Cinema Composable Image Set specification The CIS specification supports composable images, which allow interactive coloring and manipulation (such as layer on/off control, etc.), and allows scientists access to float values from within the images.

• <strong>Jupyter Notebook support</strong> Jupyter notebooks are a very flexible way for scientists to interact with Cinema databases. In conjunction with the CIS image format, this provides novel ways to analyze and present Cinema databases. Included in the cinemasci module are renderers and viewers providing reference implementations for Jupyter-based viewers.

The Cinema Ecosystem

The Cinema ecosystem consists of database specifications, writers, viewers and algorithms. Among the many use cases for Cinema are two common ones:

• If you have a scientific dataset, and you'd like to create a Cinema database, you can easily export one using the Cinema writers incorporated into ParaView and VisIt.
• You can write your own database, using your current data, by referencing one of the database specifications in this repository.

Cinema Open Source releases

All of the Cinema open source code can be found under the cinemascience group at https://github.com/cinemascience. This repository collects the officially released code. Other repositories under the cinemascience organization are considered experimental, and are not supported as production code by the Cinema team, but are viable projects dedicated either to experimental applications, algorithms or viewers.

Cinema Release

The Cinema toolkit consists of the following directories and repositories. Version information, test plans and documentation can be found in the individual repositories:

• specs/, a set of specifications for Cinema databases
• cinemasci/
  • the cinemasci module, which now contains the bulk of the components, including viewers.

Checking out this toolkit

Please note that this repository contains submodules, so you will have to update those modules after cloning:

git submodule update --init --recursive

Cinema mailing list

Please mail cinema-info@lanl.gov with any questions.

Acknowledgements

Cinema is a research project managed by the Data Science at Scale Team team at Los Alamos National Laboratory.