He asked if ioda converters are fast enough to process such data in an operational setting. Jianjun then pointed out that, in practice, radiance data often has many locations and hundreds or even thousands of channels. And, they are looking into tools that will allow users to specify these translations in yaml files. The JEDI team is working with the OBS team to define standard naming conventions for ioda. All converters need to map variable names from those that are in the file to ioda naming conventions. Steve acknowledged that this is a common occurrence, though he did not focus on it today. Jianjun asked about the case when the input and output variables for the converters are different. When Steve completed his presentation the floor was opened for questions. But, there is at least one attribute that is required for all data: make sure you specify the units for each Variable. The JEDI team is working with the OBS team to define required attributes for different data types.
#Define ioda how to#
Just one note of clarification here: on slide 13 Steve described how to attach attributes to variables. The slides have detailed information and examples. Then Steve H walked us through the tutorial presented on the following slides: Please let Yannick know if you wish to participate, if you haven't already.
#Define ioda code#
Yannick also reminded everyone of two upcoming code sprints, one for writing tests (Sept 21-25) and one for writing documentation (Oct 5-9). An example is the MPI cleaning one that was just merged but there are more coming, in particular one that will require changes for anyone using 4Denvar (this includes at least LFRic and MPAS, maybe others?). Or, if you are unable to address it soon, please ask someone from your group to address it. If you are asked to review a PR please do it in a timely manner. First, a reminder that in preparation for the release there will continue to be many pull requests that may require action from you, for example updating models. Yannick opened the meeting by announcing the focused topic discussion of the day: A tutorial on writing a ioda-converter that is compatible with the new ioda-engines.īefore proceeding to the tutorial, Yannick had a few announcements.
0 kommentar(er)
