[Nek5000-users] nhis

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

Hi,

yes, after setting nhis=npts (number of locations in hpts.in) instead of nhis=>npts/nproc, the results in hpts.out where what I expected, no ‘messed up’ values.

Thanks.


On Dec 5, 2013, at 2:16 PM, nek5000-users at lists.mcs.anl.gov wrote:

> 
> Hi,
> 
> I have found that if you set lhis to npts/np you get data that's messed up and out of order, BUT if you set lhis to npts it works fine. I think there is a bug in hpts (see my email to the users group on Nov 13 and copied below).
> 
> 
> However, I have some pretty big grids, so I need to use the parallelization feature (i.e., set lhis to npts/np, here npts = total number of interpolation 
> points and np =number of processors). However, then the data output to hpts.out  is jumbled up. If I instead set lhis to the total number of points, the output data is
> fine. I suspect there is a bug somewhere in  hpts such that it mixes up the data.
> 
> Cheers,
> 
> Janet
> 
> 
> 
> ------------------------------
> 
> Message: 3
> Date: Wed, 4 Dec 2013 15:11:29 -0600
> From: nek5000-users at lists.mcs.anl.gov
> To: nek5000-users at lists.mcs.anl.gov
> Subject: Re: [Nek5000-users] nhis
> Message-ID:
>         <mailman.7110.1386191513.3278.nek5000-users at lists.mcs.anl.gov>
> Content-Type: text/plain; charset="windows-1252"
> 
> Hi,
> 
> nhis in SIZE is the number of points, per processor, that can be read in.
> So, you will want to be sure that the total number of points in hpts.in =
> nhis*np.
> 
> If this is the case and you are still getting strange values in hpts.out,
> we can look at other possible problems.
> 
> Thanks
> Katie
> 
> 
> On Wed, Dec 4, 2013 at 11:41 AM, <nek5000-users at lists.mcs.anl.gov> wrote:
> 
> > " <nek5000-users at lists.mcs.anl.gov>
> > Mime-Version: 1.0 (Mac OS X Mail 7.0 \(1822\))
> > X-Mailer: Apple Mail (2.1822)
> >
> > Hi,
> >
> > I would like to extract the time-averaged values from a avgcase0.f00001
> > file of, for example, the velocity components at some specific locations.
> >
> > To do that, i set nek5000 in "post-processing mode? by setting nsteps=0,
> > read the file in the userchck and call hpts().
> > I need a large number of locations I want to interpolate at (stored in
> > hpts.in), much larger than the default nhis=100 in SIZE.
> > I am finding spurious values in the hpts.out file at some locations. These
> > spurious values and their locations change when I change nhis.
> >
> > I am doing something wrong?. There is a better way to do that?.
> >
> > Thanks.
> > _______________________________________________
> > Nek5000-users mailing list
> > Nek5000-users at lists.mcs.anl.gov
> > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
> 
> 
> 
> 
> -- 
> Janet Scheel
> Assistant Professor of Physics
> Occidental College
> 1600 Campus Road, M21
> Los Angeles, CA 90041
> (323) 259-2777
> jscheel at oxy.edu
> _______________________________________________
> Nek5000-users mailing list
> 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/20131205/7185e674/attachment.html>