[mesa-users] RAM issues in sdb_test
Bill Paxton
paxton at kitp.ucsb.edu
Thu Jun 27 12:29:06 EDT 2013
Hi Sutirtha,
I hope you are willing to do some digging on this one since there's nothing obvious for a quick fix.
A good place to start would be to try testing the op_mono opacities separately from mesa/star.
You could edit the mesa/kap/test/src/sample_kap.f file to call the op_mono kap/public/kap_lib routines:
load_op_mono_data
get_op_mono_params
op_mono_get_kap
Let me know what happens.
Good luck,
Bill
On Jun 27, 2013, at 9:14 AM, sutirtha sengupta wrote:
> Hello,
> I am using MESA (with SDK) on a 64-bit quad-core desktop with 8GB of RAM in Ubuntu(12.04) framework. I want to test the latest updates to the diffusion routines (including radiative levitation). I downloaded the OPCD_3.3 data file from the OP website and installed it to create the data files (in the mono directory) required by mesa for calculating radiative accelerations. But I run out of memory (RAM) as soon as mesastar loads the data into memory i.e. after the screen shows 'reading OP cache file .../data/kap_data/op_mono_cach.bin' - I checked the op_mono_cache.bin file created during the test run (with the sdb test suite in MESA version 4936) - it is 869MB in size and I have 7GB of RAM free. I also tried running on a single thread but still it uses up the whole RAM (and I can hardly kill the process!). I did not make any changes to the test suite (sdb) and am using the same inlist (inlist_sdb). Any ideas how to resolve this issue would help.
> Thanks,
> Sutirtha.
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Windows:
>
> Build for Windows Store.
>
> http://p.sf.net/sfu/windows-dev2dev_______________________________________________
> mesa-users mailing list
> mesa-users at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/mesa-users
More information about the Mesa-users
mailing list