Dresden 2026 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
AKPIK: Arbeitskreis Physik, moderne Informationstechnologie und Künstliche Intelligenz
AKPIK 6: AI Methods for Physics and Materials Science
AKPIK 6.5: Talk
Thursday, March 12, 2026, 17:45–18:00, BEY/0127
FAIR and Flexible Workflow Support within the NOMAD Infrastructure — •J.F. Rudzinski1, T. Bereau2, S. Botti3, E.B. Boydas1, N. Daelman1, L. Himanen1, S. Kapoor1, A.N. Ladines1, J.A. Márquez1, B. Mohr1, H. Näsström1, and Fairmat Team1 — 1CSMB, HU Berlin — 2ITP, Heidelberg Uni. — 3RC-FEMS, Ruhr Uni. Bochum
NOMAD [nomad-lab.eu] [1, 2] is an open-source, community-driven research data infrastructure designed for modern physics. It provides FAIR-compliant storage, management, and analysis for diverse computational and experimental materials science data, and its modular, plugin-based architecture enables low-barrier extensions for adjacent and interdisciplinary domains. Here we present NOMAD*s workflow capabilities as a foundation for scalable and AI-ready data pipelines. A general workflow schema supports both standardized and custom workflows that record detailed provenance and link heterogeneous data streams. Standardized workflows enable powerful search, visualization, and automation features, while custom workflows support agile, project-specific digitalization. Workflow entries can be created via Python-based plugins, a YAML workflow specification, or the NOMAD ELN interface, ensuring accessibility for researchers with varying technical backgrounds. Combined with a toolkit for high-throughput interfacing, NOMAD provides a robust and sustainable digital infrastructure across physics subdisciplines.
[1] Scheidgen, M. et al., JOSS 8, 5388 (2023).
[2] Scheffler, M. et al., Nature 604, 635-642 (2022).
Keywords: FAIR data; research data management; scientific workflows; data provenance; AI-ready datasets