Speaker
Description
Next-generation photon sources face increasing demands for handling high data throughput and highly diverse experimental methodologies. We have developed a modular workflow management system (WMS) designed to streamline data processing pipelines for multi-disciplinary beamlines.
The system is built upon our "Daisy" software framework, which adopts architectural patterns inspired by established frameworks like Gaudi and Mantid. This foundation allows for a clear separation between algorithm development and execution logic. Key features of the WMS include a hierarchical encapsulation mechanism that enables scientists to share and reuse specialized methods, and an internal engine that supports graphical orchestration and distributed task execution. Furthermore, by modeling the full lifecycle of data elements, the system enhances the reproducibility and verifiability of scientific results.
The framework's utility has been demonstrated through pilot implementations in automated data reduction and analysis for typical synchrotron experiments. At NOBUGS 2026, we hope to share our practical experiences in building this system and discuss how such a workflow-driven approach can better support user groups in managing complex experimental data.
