isis autoreduction

[Anders Markvardsen]

Adding a placeholder ticket isis autoreduction for this release which as a minimum will involve getting new hardware to run autoreduction for let and working with computing group member(s) to getting large data to fit into existing archiving public access structure

[Nick Draper]

Bulk move of tickets out of triage (new) to assigned at the introduction of the triage state

[Anders Markvardsen]

Start to log work which is done on autoreduction that may be useful for others or myself to read later when setting or working with autoreduction.

1. Script modified to handle multiple path locations for raw files and that there is a fallover mechanism if one is not working, uploaded to ​https://github.com/mantidproject/autoreduce/tree/master/ISISPostProcessRPM

2. Be aware when sending jsons that it will not handle text like "run_number": 00015997. It needs to be either "run_number": 15997 or "run_number": “00015997”

[Anders Markvardsen]

As far as I remember most of work for this task was done prior and during ISIS cycle 13_5 in time for LET scientists to try it out (at the time this was a high critical task in order to provide this service to LET and also get started using new hardware for this, machine now with the name autoreduce.isis.cclrc.ac.uk) and then in anger for the following cycle.

To tester, ISIS ONLY:

1. check files in archive location:
   isis\inst$\NDXLET\Instrument\data\cycle_13_5\autoreduced

2. check files in archive location:
   isis\inst$\NDXLET\Instrument\data\cycle_14_1\autoreduced

3. during this release an isis rpm build was set up, ​https://github.com/mantidproject/autoreduce/tree/master/ISISPostProcessRPM, optionally you may take a look at this branch

PS: Note it has been observed that for LET they produced an order of magnitude more reduced data than raw data. This has been noted with the instrument scientists and ISIS staff involved with storage and the archive

[Stuart Campbell]

This ticket has been transferred to github issue ​9799