Re: [mesa-users] MESA with very large network.

[Robert F. <rjf...@as...>]

Hi

I'm afraid its a known limitation of mesa, there is a maximum sized problem
based on the number of isotopes**2 * number of zones < some limit.
Basically we use a dense storage system for the matrix in mesa, which
scales in size as isotopes**2 * number of zones, and eventually with a big
enough problem we overflow the 4-byte integer used to index the array which
gives you the segmentation fault.

For ~400 isotopes you can have a maximum of ~1000 zones. So i'm afraid you
cant run a 1700 isotope network on a star in mesa. You best bet is to stick
with a few hundred isotope network and do some sort of post processing step
afterwards if you need the extra isotopes.

Rob


On Tue, Apr 11, 2017 at 1:10 AM, Byeong-Chan Park <pb...@gm...> wrote:

> Dear
>
> Hello. I'm a graduate student using mesa8118. I'm trying to create
> pre-supernova model based on the massive_rotating test suite files with
> very large network.
>
> When I tried to run the code with rp305.net, there is no problem. Thus I
> tried to run same code with very large network including 1700 isotopes. The
> format of net file was based on the rp.net.
>
> However, the termincal showed segmentation fault problem. I attached image
> file including the terminal log (log.pdf) because the terminal was
> frozen......
>
> It seems that the log indicates there is some problem in a subroutine of
> 'hydro_mtx.f90' file, but I cannot find the way how I revise this
> subroutine.
>
> I need your help.
>
> Reagrds
> Park
>
> ------------------------------------------------------------
> ------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> _______________________________________________
> mesa-users mailing list
> mes...@li...
> https://lists.sourceforge.net/lists/listinfo/mesa-users
>
>