view article
Figure 3
Data workflow in OnDA. Each worker node collects event data from a source, which can be a shared memory server, a file or a network data stream. It then calls the extract_data function, which is imported from the Data Extraction Layer, to recover useful information from the event data (a detector readout, an instrument reading) and makes it available as a class property. The worker then calls the process function, imported from the Processing Layer, to process the information, and stores the results in a Python dictionary called results_dict. This dictionary is then sent to the master node using the MPI framework. Upon receiving the data, the master note calls the collect function, which carries out further processing, and stores what needs to be sent to the GUI in another Python dictionary called collected_data.

Journal logoJOURNAL OF
APPLIED
CRYSTALLOGRAPHY
ISSN: 1600-5767
Volume 49| Part 3| June 2016| Pages 1073-1080
doi:10.1107/S1600576716007469
Follow J. Appl. Cryst.
Sign up for e-alerts
E-alerts
Follow J. Appl. Cryst. on Twitter
Twitter
Follow us on facebook
Facebook
Sign up for RSS feeds
RSS