Hi Ken,<br><br>I think it's more than just a file contention issue. On hopper@nersc I did set DVS_MAXNODES to 14 and that helped out a lot. Without that set before I was able to run with 480 processes accessing the same data file (the <span>17*768*1152 with 324 time steps data set) but with the "bad" one that was </span><span>768*1152 with 9855 time steps</span> I had problems with just 24 processes.<br>
<br>I have some things which I want to try out but I think you're right that using a parallel netcdf library should help a lot, if it doesn't cause conflicts.<br><br>Thanks,<br>Andy<br><br><div class="gmail_quote">
On Wed, Feb 6, 2013 at 5:20 PM, Moreland, Kenneth <span dir="ltr"><<a href="mailto:kmorel@sandia.gov" target="_blank">kmorel@sandia.gov</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div style="font-size:14px;font-family:Calibri,sans-serif;word-wrap:break-word">
<div>
<div>
<div>This does not surprise me. The current version of the netCDF reader only uses the basic interface for accessing files, which is basically a serial interface. You are probably getting a lot of file request contention.</div>
<div><br>
</div>
<div>At the time I wrote the netCDF reader, parallel versions were just coming online. I think it would be relatively straightforward to update the reader to use collective parallel calls from a parallel netCDF library. Unfortunately, I have lost track on
the status of the parallel netCDF library and file formats. Last I looked, there were actually two parallel netCDF libraries and formats. One version directly added collective parallel calls to the library. The other changed the format to use hdf5 under
the covers and use the parallel calls therein. These two libraries use different formats for the files and I don't think are compatible with each other. Also, it might be the case for one or both libraries that you cannot read the data in parallel if it
was not written in parallel or written in an older version of netCDF.</div>
<div><br>
</div>
<div>-Ken</div>
<div>
<div><br>
</div>
</div>
</div>
</div>
<span>
<div style="border-right:medium none;padding-right:0in;padding-left:0in;padding-top:3pt;text-align:left;font-size:11pt;border-bottom:medium none;font-family:Calibri;border-top:#b5c4df 1pt solid;padding-bottom:0in;border-left:medium none">
<span style="font-weight:bold">From: </span>Andy Bauer <<a href="mailto:andy.bauer@kitware.com" target="_blank">andy.bauer@kitware.com</a>><br>
<span style="font-weight:bold">Date: </span>Wednesday, February 6, 2013 10:38 AM<br>
<span style="font-weight:bold">To: </span>"<a href="mailto:paraview@paraview.org" target="_blank">paraview@paraview.org</a>" <<a href="mailto:paraview@paraview.org" target="_blank">paraview@paraview.org</a>>, Kenneth Moreland <<a href="mailto:kmorel@sandia.gov" target="_blank">kmorel@sandia.gov</a>><br>
<span style="font-weight:bold">Subject: </span>[EXTERNAL] vtkNetCDFCFReader parallel performance<br>
</div><div><div class="h5">
<div><br>
</div>
<blockquote style="BORDER-LEFT:#b5c4df 5 solid;PADDING:0 0 0 5;MARGIN:0 0 0 5">
<div>
<div>Hi Ken,<br>
<br>
I'm having some performance issues with a fairly large NetCDF file using the vtkNetCDFCFReader. The dimensions of it are 768 lat, 1152 lon and 9855 time steps (no elevation dimension). It has one float variable with these dimensions -- pr(time, lat, lon). This
results in a file around 33 GB. I'm running on hopper and for small amounts of processes (at most 24 which is the number of cores per node) and the run time seems to increase dramatically as I add more processes. The tests I did read in the first 2 time steps
and did nothing else. The results are below but weren't done too rigorously:<br>
<br>
numprocs -- time<br>
1 -- 1:22<br>
2 -- 1:52<br>
4 -- 7:52<br>
8 -- 5:34<br>
16 -- 10:46<br>
22 -- 10:37<br>
24 -- didn't complete on hopper's "regular" node with 32 GB of memory but I was able to run it in a reasonable amount of time on hopper's big memory nodes with 64 GB of memory.<br>
<br>
I have the data in a reasonable place on hopper. I'm still playing around with settings (things get a bit better if I set DVS_MAXNODES --
<a href="http://www.nersc.gov/users/computational-systems/hopper/performance-and-optimization/hopperdvs/" target="_blank">
http://www.nersc.gov/users/computational-systems/hopper/performance-and-optimization/hopperdvs/</a>) but this seems a bit weird as I'm not having any problems like this on a data set that has spatial dimensions of 17*768*1152 with 324 time steps.<br>
<br>
Any quick thoughts on this? I'm still investigating but was hoping you could point out if I'm doing anything stupid.<br>
<br>
Thanks,<br>
Andy<br>
<br>
<br>
</div>
</div>
</blockquote>
</div></div></span>
</div>
</blockquote></div><br>