12.02.2024 10:00 – 10:45
New mode of ingesting metadata at P08
Since a few weeks another mode to ingest meta data has been developed and is deployed at P08 and can be configured in several ways.
Jan presented how metadat was previously ingested using essentially a copy_map (take all meta data from nexus file) to filter and rearrange (order) information of the data. With the new mode, by adding a few macros, one copies less information from the nexus file otherwise the infrastructure is the same. Grouping is achieved by introducing reduction algorithms like looking for minimum, maximum, min and max, forming the average value, etc. Use dictionaries to store units, but one can adopt commands to needs.
With the new ingestion mode, one can group related scans to one datasets, e.g. if one wants to collect more statistics, i.e. in cases where a large commonality exist between a scan and its preceding scan. The idea was to reduce the amount of datasets.
Another mode is automatic grouping: an update in SciCat is done every scan, it can therefore create a lot of history meta data as after each scan. Deployment to be seen.
CONCLUSIONS
Technically this new mode works, was tested OK.
But dedicate only time to setting up a good configuration of grouping datasets once it is clear what information is stored. Earliest once they are back in operation – which is unclear due to a broken crane in the hall.
Participants
1.Jan K
2.Martin G
3.Anton B
4.Dima K
5.Sabrina S
6.Anjali A
7.Igor K
8.Fabiano L
9Udai S
10.RH