1 6 8 H I G H L I G H T S 2 0 2 3 I

E N A B L I N G T E C H N O L O G I E S

This year s Enabling Technologies chapter clearly emphasises that the ESRF is actively addressing the challenges posed by the increasing amounts of data generated and collected thanks to the Extremely Brilliant Source (EBS), as well as the introduction of artificial intelligence (AI) and machine-learning (ML) techniques to optimise the accelerator and experiments. The development and deployment of the new beamline control system (BLISS) on most beamlines has enabled the integration of new functionalities to enable experiments to make the most of the unique capabilities of the EBS. The decision to develop BLISS was taken at the right time so that beamlines can fully profit from the EBS.

The first article, on page 170 sheds light on the ongoing developments at the ESRF, where the imperative to navigate with huge volumes of data has required a profound transformation in dataflow management. The new ESRF Workflow System (EWOKS) is a generic meta-workflow solution developed for automating data processing and making it repeatable in full accordance with the principle of findable, accessible, interoperable and reusable (FAIR) data management. Furthermore, the possibility of strong interactions between EWOKS and BLISS illustrates the full power of BLISS, offering users a more integrated environment than other BLISS alternatives.

The second and third articles showcase the rise of AI/ML solutions for both online and offline data analysis. Pithan et al. (page 172) demonstrate the benefit of using AI technologies to enhance the effectiveness of experiments, supporting users and staff in on-the-fly data analysis to dynamically adapt experiment parameters, closing the loop with using the TANGO protocol and servers and storing the processed results in ICAT, the ESRF metadata catalogue. Meunier and Burtin (page 174) highlight the use of AI technologies in the context of data analysis that has been developed to optimise the accelerator operation.

Through these articles, the ESRF sees AI/ML technologies playing a role in optimising beamtime usage, data production, data workflows, data management and machine parameters, which all ultimately help users to produce scientific publications.