How warm_start works?
Page 1 of 1

Author:  dhsong [ Tue Jul 03, 2018 1:52 pm ]
Post subject:  How warm_start works?

Hi there,

I am doing some 3D simulation and I need to break my job to several parts due to CPU walltime limits. I think I can use the warm_start option in galprop to help me with that. But I am not sure how it works exactly. According to the documents:

warm_start   = 0 read in nuclei file and continue run
Flag to indicate the run is to be started with results output from previous run with output gcr full =1. Useful for long runs which have to be broken up due to CPU time limitations. NB the galdef file should be the same as for the first run, apart from this parameter; the only possible differences should be the values of end timestep and timestep repeat2, and the generation of gamma-rays etc. The grid and selected nuclei, electrons etc. must be the same.

From my understand I can break my job to parts by set different end timesteps. For example if I need to propagate from timestep 1e+9 yr to 1e+2 yr, I can run the first job with:

output_gcr_full = 1
warm_start = 0
start_timestep = 1e+9
end_timestep = 1e+5

And for the second run,

output_gcr_full = 1
warm_start = 1
start_timestep = 1e+9
end_timestep = 1e+2

Then galprop would continue the first run and pick up the propagation from timestep = 1e+5 yr.

However I did some test runs and only found that in the second run with such configurations, galprop still started the propagation from 1e+9 yr. I guess I didn't figure out how to use the warm_start properly.

So how warm_start works exactly? what should I do to break up my job?


Page 1 of 1 All times are UTC - 8 hours [ DST ]
Powered by phpBB® Forum Software © phpBB Group