[FLASH-USERS] BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES , EXIT CODE 9

Klaus Weide klaus at flash.uchicago.edu
Fri Sep 28 21:33:36 EDT 2018


On Fri, 28 Sep 2018, g.granda at irya.unam.mx wrote:

> Hello,
> I have been using Flash 4.5 to run simulations of thermal instability for a
> while. Everything was fine until
> I tried to increase the resolution of the uniform grid used for previous
> simulation. I don't know what is the problem since the error message is not
> that informative, but the simulation crash right after enetering the evolution
> loop.
> 
> Could you please help me?

Well the most likely assumption is that there is some limit on the 
amount of memory you can use on that system, and you have made your block 
size so large that the total memory that FLASH needs exceeds the limit.

You are using on each MPI rank 136*136*136 cells (this includes guard 
cells). Multiply by number of variables times 8 (sizeof double precision),
and you the the number of bytes you need for solution storage (UNK) alone.
Maybe you just cannot use that much.

Not sure why this problem would only manifest itself in the first step of 
the evolution phase.

Klaus



More information about the flash-users mailing list