Here I have 400 input files(they are different from each other in some details) to run，so it’s troublesome if I choose to run them manually(open them with flair and click the “RUN” indepently ), I’m wondering whether there is an efficient way to batch run them or not.
I have tried to use “Add” to add some input files to “RUN” of flair in order to run them at once , but I found that the added files become the same automaticlly, so is there any other way to batch run plenty of input files?
You can run them from the command line with
flair --run runname
or even a pattern as
flair --run "myrun*"
If you are running them locally is better to select as a submission system the tsp (=taskspooler) option in flair,
which behaves like a very primitive batch scheduler.
You select all the runs Ctrl-A and press Start
You can configure the tsp to run up to a certain number of concurrent jobs (e.g. to the number of cores)
You need to install the additional linux package of task-spooler
Thank you very much for your quickly reply, I have trid to use tsp option as you taught before , and I set “tsp -S 5” in the terminal. Howerver, I found the flair can just run the first five files of the 400 input files successfully，the others display the status of “TIMED OUT” and can’t run as expected
So could please teach me how to solve this problem
By setting “tsp -S 5” you are indicating to run up to 5 jobs at the same time. When one job finishes the next one in the line will take over. However flair is completely agnostic of the launching program. It check only the output of fluka to find out what happens in the run. Since the runs are in the waiting queue flair didn’t find any output and it reported that the attaching process timed out.
What you can do is reselect the runs and click Attach, to find out which runs are going on and which ones have finished
Thank you very much for your helpful reply. I have already solved my problem