Speaker
Description
Abstract. Many X-ray based techniques allow for simultaneous acquisition, e.g. XRF, XRD, SAXS, WAXS, XBIC, XBIV, and
many more. Although previously hardly every utilised, it is nowadays becoming the standard. Now that many technical challenges
are solved the next challenge lays in the analysis of the acquired data.
INTRODUCTION
Adding the possibility of multimodal data acquisition is a must for any beamline. The current challenges are not only
the amount of data and computational resources needed, but also the know-how. This because many users are not
specialists any more in the x-ray field but in another research field. To make the best of the data, data analysis often
is/was outsourced to beamline personnel.
ARCHITECTURE
The main architecture is based on micro services and distributed on as many compute nodes as required for the task at
hand. All acquired data is accompanied by metadata that allows for fully automated data analysis for every available
measurement modality. The default data pipeline constitutes out of 3 components, a master, worker, and data sorter.
A task distributor and a sink complete the default standard blocks.
AIM
The goal of the data pipelines is to provide processed data asap to the visiting scientist and control system. The delay
between data acquisition and analysed data is dependant on several components. Software stack of the used detector,
streaming or file based, and the methodology and the required computational power. The analysed data is provided to
the user in 3 forms:
•
•
•
Queryable database;
HDF5 file;
Visualised in an online data viewer.
ACKNOWLEDGMENTS
We acknowledge DESY (Hamburg, Germany), a member of the Helmholtz Association HGF, for the provision of
experimental facilities. Parts of this research were carried out at the PETRA III P06 Beamline.
| Abstract publication | I agree that the abstract will be published on the web site |
|---|
