MatFlow: A Python API for computational Materials Science workflows
Stavrina Dimosthenous, Aman Goel
Henry Royce Institute, University of Manchester · University of Manchester
2025-06-17 · 09:35 BST
We present MatFlow and HPCFlow, collectively referred to as MatFlow, (GitHub Repo), a workflow engine for running computational tasks locally and on high-performance computing (HPC) clusters.
We will discuss the capabilities of the MatFlow command-line interface, the MatFlow Python API, built-in task schemas which integrate Materials Science computational tools, future work, and use cases.
Stavrina and Aman will be delivering this talk on behalf of Guy Bowker, Adam Plowman
, João Quinta da Fonseca
and the MatFlow development team.
MatFlow was originally developed by Adam Plowman, supported by the EPSRC grant EP/R001715/1, LightForm, and is actively supported by researchers at the University of Manchester’s CLARI group and collaborators, and at UKAEA. For an example of usage, see Plowman et al. 2023 (DOI) [version 1].