Hi all,
Since a few weeks I am co-advising a PhD student in Leeds and I’m currently trying the help her set up running the UM and processing the output. However, I am a bit confused by the typical workflow for running and analysing the UM from outside the Met Office (where I am based), and was wondering whether anyone here might be able to help?
The student has recently followed the NCAS introduction to the UM course and as far as I am aware has an account on the ARCHER2 and on JASMIN. For the moment, it appears that she would be setting up (using the CYLC/Rose GUI) and running the UM on ARCHER2. However, iris does not seem to be available on ARCHER2 and it seems she will have to do any data processing on JASMIN.
Would you know whether it is possible at all to run the UM directly on JASMIN, so we can just stick to a single platform to run the model and process the data, rather than having to copy files back and forth? If not, I wondered how people outside the office typically manage their workflow in terms of accessing the Science Repository from both ARCHER2 and JASMIN and having access to their data across these two HPCs. Is it a matter of archiving data to MASS from ARCHER2 and then retrieving data from MASS again on JASMIN to do any post-processing?
Thanks a lot in advance for any tips or suggestions!
Kwinten