NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.
To learn about what's going on, see About the Archive Site.
NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.
To learn about what's going on, see About the Archive Site.
On 10/17/2014 07:33 AM, Samrat Rao wrote:
Initially i tried with pnetcdf, wrote a few variables, but found that the format was CDF-5 which 'normal' netcdf would not read.
Yes, sorry about that. We're working on improving CDF-5 support, but it's not quite there yet.
I also need to read some bits of netcdf data in Matlab, so i thought of sticking to the usual netcdf-4 compiled for parallel io. It is also likely that i will have to share my workload with others in my group and/or leave the code for future people to work on. Does matlab read cdf-5 files? So i preferred the usual netcdf. Rob, i hope you are not annoyed.
sure, no problem. you need to tackle the data decomposition problem first anyway.
But most of the above is for another day. Currently i am stuck elsewhere. With a less no of processors, 216, the single netcdf file gets created (i create i single netcdf file for each variable), but for anything above that i get these errors: NetCDF: Bad chunk sizes. Not sure where these errors come from.
that's an unusual error. If you simplify this a bit and have 216 processors writing a single variable, will you still get this error?
Then i shifted to dumping outputs from each processor in simple binary --- this works till about 1500 processors. Above this number the code gets stuck and eventually aborts.
gets stuck how?
This issue is not new. My colleague too had problems with running his code on 1500+ procs. Today i came to know that opening a large number of files (each proc writes 1 file) can overwhelm the system --- solving this requires more than rudimentary techniques of writing --- or understanding the system's inherent parameters/bottlenecks.
well, it's not quite that simple. some poorly designed file systems require one file per process. GPFS, for a counter-example, will spend quite a long time on file creation if you try to create one file per process.
So netcdf is probably out of bounds for now --- will try again if the simple binary write from each processor gets sorted out. Does anyone have any suggestion?
yes, you need to re-think your data decomposition. Instead of one file per process, partition your array. you can do this face-wise, pencil-wise, subcube-wise... really depends on your science demands
(for example, a volume renderer will not do well with face-wise decomposition: it requires a 3d region for the math to work out)
Also, ensure you are using MPI-IO, and collective MPI-IO at that. ==rob
In Thanks, Samrat. On Thu, Oct 2, 2014 at 7:52 PM, Rob Latham <robl@xxxxxxxxxxx <mailto:robl@xxxxxxxxxxx>> wrote: On 10/02/2014 01:24 AM, Samrat Rao wrote: Thanks for your replies. I estimate that i will be requiring approx 4000 processors and a total grid resolution of 2.5 billion for my F90 code. So i need to think/understand which is better - parallel netCDF or the 'normal' one. There are a few specific nifty features in pnetcdf that can let you get really good performance, but 'normal' netCDF is a fine choice, too. Right now I do not know how to use parallel-netCDF. It's almost as simple as replacing every 'nf' call with 'nfmpi' but you will be just fine if you stick with UCAR netCDF-4 Secondly, i hope that the netCDF-4 files created by either parallel netCDF or the 'normal' one are mutually compatible. For analysis I will be extracting data using the usual netCDF library, so in case i use parallel-netCDF then there should be no inter-compatibility issues. For truly large variables, parallel-netcdf introduced, with some consultation from the UCAR folks, a 'CDF-5' file format. You have to request it explicitly, and then in that one case you would have a pnetcdf file that netcdf tools would not understand. In all other cases, we work hard to keep pnetcdf and "classic" netcdf compatible. UCAR NetCDF has the option for an HDF5-based backend -- and in fact it's not an option if you want parallel I/O with NetCDF-4 -- is not compatible with parallel-netcdf. By now, your analysis tools surely are updated to understand the new HDF5-based backend? I suppose it's possible you've got some 6 year old analysis tool that does not understand NetCDF-4's HDF5-based file format. Parallel-netcdf would allow you to simulate with parallel i/o and produce a classic netCDF file. But I would be shocked and a little bit angry if that was actually a good reason to use parallel-netcdf in 2014. ==rob -- Rob Latham Mathematics and Computer Science Division Argonne National Lab, IL USA -- Samrat Rao Research Associate Engineering Mechanics Unit Jawaharlal Centre for Advanced Scientific Research Bangalore - 560064, India
-- Rob Latham Mathematics and Computer Science Division Argonne National Lab, IL USA
netcdfgroup archives: