Kuibit: a community-friendly tool to analyze Cactus simulations

Thursday, February 4, 2021, 9am US central time

Slides are available as pdf, and the recordings are available in webm and mp4 format.


Analyzing Einstein Toolkit simulations can be a challenging task. Usually, the process involves a series of technical obstacles like combining data from different restarts or working with HDF5 files. In this talk, I will present kuibit, a Python3.6+ library that takes care of all these low-level details (and many other more) and that provides high-level, intuitive, representations of the data. Kuibit comes with a wide range of features that include full support for 1-3D ASCII and HDF5 grid data, time and frequency series, gravitational waves, and apparent horizons. With kuibit, users can inspect most of the content of a simulation with just a few lines of code. Importantly, kuibit is designed to be a code for the community: it is user-friendly and does not require any proprietary software to run, it has documentation and examples, and it is openly developed with emphasis on extensibility and maintainability. In the talk, I will first give a general overview of the project, then I will discuss in more details the various modules.


kuibit is available under GPLv3. kuibit follows the same design and part of the implementation details of PostCactus, code developed by Wolfgang Kastaun. This fork completely rewrites the original code, adding emphasis on documentation, testing, and extensibility. The logo contains elements designed by freepik.com.

Presenter: Gabriele Bozzola

Gabriele Bozzola

PhD student, University of Arizona. read more.

Gabriele studies the fragmentations of supermassive stars that result in binary black holes surrounded by accretion disks.