[Nek5000-users] Restart Problem

[nek5000-users at lists.mcs.anl.gov]

Sorry ---

Looking again at my_full_restart, I see that it and the routines
it calls are both very streamlined and don't expose the variables
of interest.

Typically, I would do something of the following form, borrowing
from "full_restart" in ic.f:


       call opcopy(vxt,vyt,vzt,vx,vy,vz)  ! Save vx,vy,vz

       call chcopy (initc,s80(ifile),80)  ! Load your field of interest
       call bcast  (initc,80)
       call restart       (1)

       call opcopy(u1,u2,u3,vx,vy,vz)     ! Save your field in u1-u3

       call opcopy(vx,vy,vz,vxt,vyt,vzt)  ! Restore vx,vy,vz


Be sure you declare vxt,vyt,vzt, u1,u2,u3, etc. to be
fields of size lx1*ly1*lz1*lelt, each.

I would encapsulate the above in a routine that is called
from userchk() as needed when istep=0 (for initial conditions)
and for istep=1,2 if you wish to boot up (e.g., as done in
full-restart).

Paul



On Sat, 21 Jul 2012, nek5000-users at lists.mcs.anl.gov wrote:

>
> Hi Alireza,
>
> I would suggest looking closely at the subroutine "my_full_restart"
> in the cyl_restart example .usr files.  I think from there you can
> figure out how to accomplish your goal.
>
> Please let me know if that gets you to where you need to be.
>
> If I were you, I would use outpost(), rather than outpost2().
>
> Both outpost (or outpost2) and my_full_restart are fully parallel
> so nothing special required in that respect.
>
> Paul
>
>
>
> On Sat, 21 Jul 2012, nek5000-users at lists.mcs.anl.gov wrote:
>
>> Hi Neks,
>> 
>> I have a question regarding I/O procedure. At each time and on every grid 
>> point, I integrate the components of a vector field in time which are 
>> functions of the vorticity and their own values from the previous time 
>> step. I need to store the values of these components at the last time step 
>> (preferably in a separate file) and then recall those values after 
>> restarting the simulation. I think I should use outpost2 to store the 
>> values, but I'm not sure how to open a different file other than fld file 
>> and then how to read that file to be consistent in parallel. I wonder if 
>> you have any hint.
>> 
>> 
>> Regards,
>> Alireza
>> 
>> _______________________________________________
>> Nek5000-users mailing list
>> Nek5000-users at lists.mcs.anl.gov
>> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
> _______________________________________________
> Nek5000-users mailing list
> Nek5000-users at lists.mcs.anl.gov
> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
>