[Nek5000-users] Using Cartesian grid lookup table in Nekton
nek5000-users at lists.mcs.anl.gov
nek5000-users at lists.mcs.anl.gov
Mon Jun 4 14:52:11 CDT 2018
Hi Saikat,
Nek provides routines to interpolate from its own spectral element
mesh to any other grid.
Going from any other grid to (x,y,z) [ given by entries in xm1(),ym1(),zm1()], requires an interpolator designed for that
grid.
You could, say, use some sort of piecewise polynomial or piecewise
linear interpolant --- it would be a routine that you could develop
and test as a stand-alone code on your own before putting into
your usr file.
hth,
Paul
________________________________
From: Nek5000-users <nek5000-users-bounces at lists.mcs.anl.gov> on behalf of nek5000-users at lists.mcs.anl.gov <nek5000-users at lists.mcs.anl.gov>
Sent: Monday, June 4, 2018 10:44:10 AM
To: nek5000-users
Subject: Re: [Nek5000-users] Using Cartesian grid lookup table in Nekton
Thanks Paul, I will try this.
Regarding the interpolation, do we have a subroutine that interpolates from uniform grid to GLL grid in nek5000?
Thanks,
Saikat
Saikat Mukherjee,
PhD Student,
Paul Research Group - http://www.me.vt.edu/mpaul/
Engineering Science and Mechanics,
Virginia Tech.
On Mon, Jun 4, 2018 at 7:32 AM, <nek5000-users at lists.mcs.anl.gov<mailto:nek5000-users at lists.mcs.anl.gov>> wrote:
Hi Saikat,
If it's not too big for loading onto each rank, I would do the following, which is not very parallel, but is quite effective up to modest array sizes:
subroutine usrdat2
include 'SIZE'
include 'TOTAL'
parameter (my_array_size=10000)
real my_array(my_array_size), work(my_array_size)
n = my_array_size
call rzero(my_array,n)
call rzero(work. ,n)
if (nid.eq.0) then
open (unit=49,file='myfile')
read(49,*) (my_array(k),k=1,n)
close(49)
endif
call gop(my_array,work,'+ ',n) ! Sum over all processors
return
end
I've not tested this code... just wrote it here but it should work as is.
Paul
________________________________
From: Nek5000-users <nek5000-users-bounces at lists.mcs.anl.gov<mailto:nek5000-users-bounces at lists.mcs.anl.gov>> on behalf of nek5000-users at lists.mcs.anl.gov<mailto:nek5000-users at lists.mcs.anl.gov> <nek5000-users at lists.mcs.anl.gov<mailto:nek5000-users at lists.mcs.anl.gov>>
Sent: Sunday, June 3, 2018 11:11:29 PM
To: nek5000-users
Subject: Re: [Nek5000-users] Using Cartesian grid lookup table in Nekton
Hey Paul,
Thanks for the reply. I have never really used the mpi architecture of Nek5000. I am thinking this should be a separate subroutine on the .usr file where I have to copy it onto each mpi rank?
My finite difference grid is not that big. So the interpolation into SEM points should be easy. I would like to know how to copy it onto each MPI rank.
Thanks,
Saikat
Saikat Mukherjee,
PhD Student,
Paul Research Group - http://www.me.vt.edu/mpaul/
Engineering Science and Mechanics,
Virginia Tech.
On Thu, May 31, 2018 at 7:58 AM, <nek5000-users at lists.mcs.anl.gov<mailto:nek5000-users at lists.mcs.anl.gov>> wrote:
Hi Saikat,
How large is your finite-difference look up table?
(1 million points? 1 billion? 1000?)
If it's not too large you can just copy it onto each mpi rank
and then call your own serial interpolation routine for each
SEM point, x,y,z, and store the results into an array in a
common block that you can then reference in users
Presumably you'd have to do this interpolation only one
time and then de-reference the values from the common
block on each time step with fairly low overhead.
Paul
________________________________
From: Nek5000-users <nek5000-users-bounces at lists.mcs.anl.gov<mailto:nek5000-users-bounces at lists.mcs.anl.gov>> on behalf of nek5000-users at lists.mcs.anl.gov<mailto:nek5000-users at lists.mcs.anl.gov> <nek5000-users at lists.mcs.anl.gov<mailto:nek5000-users at lists.mcs.anl.gov>>
Sent: Wednesday, May 30, 2018 2:42:14 PM
To: nek5000-users at lists.mcs.anl.gov<mailto:nek5000-users at lists.mcs.anl.gov>
Subject: Re: [Nek5000-users] Using Cartesian grid lookup table in Nekton
There is nothing spectral element specific here. You could do e.g. a trilinear interpolation but that's something you have to code yourself.
-----Original message-----
> From:nek5000-users at lists.mcs.anl.gov<mailto:From%3Anek5000-users at lists.mcs.anl.gov> <nek5000-users at lists.mcs.anl.gov<mailto:nek5000-users at lists.mcs.anl.gov>>
> Sent: Wednesday 30th May 2018 21:26
> To: nek5000-users <nek5000-users at lists.mcs.anl.gov<mailto:nek5000-users at lists.mcs.anl.gov>>
> Subject: Re: [Nek5000-users] Using Cartesian grid lookup table in Nekton
>
> The question is how to read in a discrete data set defined in a finite difference grid in ffx or forcing in the .usr file. It seems like there will be some sort of interpolation to the spectral element grid. So yes, that is my question.
>
> Thank you,
> Saikat
> <br clear="all" />Saikat Mukherjee,
> PhD Student,
> Paul Research Group - http://www.me.vt.edu/mpaul/ <http://www.me.vt.edu/mpaul/>
> Engineering Science and Mechanics,
> Virginia Tech.
>
> On Wed, May 30, 2018 at 2:48 PM, <nek5000-users at lists.mcs.anl.gov<mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov>> wrote:
> What’s your question? How to do the interpolation in your function f?
>
> On 30 May 2018, at 20:17, "nek5000-users at lists.mcs.anl.gov<mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov>" <nek5000-users at lists.mcs.anl.gov<mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov>> wrote:
>
> Hi,
>
> Thanks for the reply. My problem is f is given as a discrete data set described in finite difference grid. I want to interpolate this to ffx like you said in my .usr file. Would like to know how to proceed.
>
> Thanks,
> Saikat
> <br clear="all" />Saikat Mukherjee,
> PhD Student,
> Paul Research Group - http://www.me.vt.edu/mpaul/ <http://www.me.vt.edu/mpaul/>
> Engineering Science and Mechanics,
> Virginia Tech.
>
> On Wed, May 30, 2018 at 1:36 PM, <nek5000-users at lists.mcs.anl.gov<mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov>> wrote:
> Lets consider
>
> ffx = f(x,y,z)
>
> here f is an user function (something you have to provide) which get evaluated at x,y,z (your grid points). Interpolation is required if f is given by a discrete data set.
>
> hth
>
>
> -----Original message-----
> > From:nek5000-users at lists.mcs.anl.gov<mailto:From%3Anek5000-users at lists.mcs.anl.gov> <mailto:From%3Anek5000-users at lists.mcs.anl.gov> <nek5000-users at lists.mcs.anl.gov<mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov>>
> > Sent: Wednesday 30th May 2018 19:19
> > To: nek5000-users <nek5000-users at lists.mcs.anl.gov<mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov>>
> > Subject: Re: [Nek5000-users] Using Cartesian grid lookup table in Nekton
> >
> > Hi,
> >
> > Thanks for the reply. The look up table is in a finite difference grid. Are you saying I can still feed it to the .usr file and it should be fine?
> >
> > Thanks,
> > Saikat
> > <br clear="all" />Saikat Mukherjee,
> > PhD Student,
> > Paul Research Group - http://www.me.vt.edu/mpaul/ <http://www.me.vt.edu/mpaul/> <http://www.me.vt.edu/mpaul/ <http://www.me.vt.edu/mpaul/>>
> > Engineering Science and Mechanics,
> > Virginia Tech.
> >
> > On Sat, May 26, 2018 at 10:54 PM, <nek5000-users at lists.mcs.anl.gov<mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov<mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov>>> wrote:
> > Hi Saikat,
> >
> > You can always do the reading from the look up table from .usr file, at once then feed the data to your ffx ffy and ffz through a common block.
> >
> > On Thu, May 24, 2018 at 10:55 PM, <nek5000-users at lists.mcs.anl.gov<mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov<mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov>>> wrote:
> > Hey all,
> >
> > Just following up on the query. Would appreciate any input.
> >
> > Thanks,
> > Saikat
> > <br clear="all" />Saikat Mukherjee,
> > PhD Student,
> > Paul Research Group - http://www.me.vt.edu/mpaul/ <http://www.me.vt.edu/mpaul/> <http://www.me.vt.edu/mpaul/ <http://www.me.vt.edu/mpaul/>>
> > Engineering Science and Mechanics,
> > Virginia Tech.
> >
> > On Tue, May 22, 2018 at 4:47 PM, Saikat Mukherjee <msaikat6 at vt.edu<mailto:msaikat6 at vt.edu> <mailto:msaikat6 at vt.edu> <mailto:msaikat6 at vt.edu<mailto:msaikat6 at vt.edu> <mailto:msaikat6 at vt.edu>>> wrote:
> > Hey Neks,
> >
> > I have a forcing function which is in the form of a look up table in X-Y cartesian grid. I would like to use this forcing function in the 2-D fluid momentum equation that Nekton solves. Is there a way to do this in the .usr file or elsewhere?
> >
> > Thanks,
> > Saikat
> >
> >
> > <br clear="all" />Saikat Mukherjee,
> > PhD Student,
> > Paul Research Group - http://www.me.vt.edu/mpaul/ <http://www.me.vt.edu/mpaul/> <http://www.me.vt.edu/mpaul/ <http://www.me.vt.edu/mpaul/>>
> > Engineering Science and Mechanics,
> > Virginia Tech.
> >
> >
> > _______________________________________________
> > Nek5000-users mailing list
> > Nek5000-users at lists.mcs.anl.gov<mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov<mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov>>
> > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users> <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users>>
> >
> >
> >
> > _______________________________________________
> > Nek5000-users mailing list
> > Nek5000-users at lists.mcs.anl.gov<mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov<mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov>>
> > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users> <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users>>
> >
> >
> > _______________________________________________
> > Nek5000-users mailing list
> > Nek5000-users at lists.mcs.anl.gov<mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov>
> > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users>
> _______________________________________________
> Nek5000-users mailing list
> Nek5000-users at lists.mcs.anl.gov<mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov>
> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users>
>
> _______________________________________________
> Nek5000-users mailing list
> Nek5000-users at lists.mcs.anl.gov<mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov>
> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users>
>
> _______________________________________________
> Nek5000-users mailing list
> Nek5000-users at lists.mcs.anl.gov<mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov>
> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users>
>
>
> _______________________________________________
> Nek5000-users mailing list
> Nek5000-users at lists.mcs.anl.gov<mailto: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<mailto: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<mailto: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<mailto:Nek5000-users at lists.mcs.anl.gov>
https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/nek5000-users/attachments/20180604/615fdf8f/attachment-0001.html>
More information about the Nek5000-users
mailing list