Open CJ-Wright opened 9 years ago
Ultimately, we have in mind an "analysis store," akin to metadatastore + filestore for analysis results, capturing various intermediate steps between raw data and a published paper.
As a first step, getting a second filestore running at each beamline is relatively easy. There will be a second set of mongo collections. (Note: we might ultimately make several sets, reflecting the climate-science jargon for different "levels" of processed data. But, to start, Level 0 and Level Nonzero is sufficient.)
To interface with these collections, the filestore module with its various functions will have to be redesigned into a Filestore
object with various methods. Then users can juggle more than one Filestore
in a given session. That will be the main effort involved in this issue -- probably two days' work.
Data which has been analyzed or simulated should be separated from its generating data. As per the discussion from the NSLS-II users meeting with the cryo-geology scientists.