[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[netCDF #UUS-259413]: pnetcdf and netcdf

Dennis,

I was confused to until I got the pnetcdf tests working on spock
yesterday.

I think the pnetcdf code should stay in libsrc4, because it can only be
built with netCDF-4 enabled.  This documentation Ed wrote about it
cleared things up for me:

  http://www.unidata.ucar.edu/software/netcdf/docs/build_parallel.html

The last paragraph states that --enable-pnetcdf is an add-on to the
netCDF-4 parallel I/O, and even requires an HDF5 built with
--enable-parallel to build correctly.  That's what I found when I
finally succeeded in building pnetcdf support.

The --enable-pnetcdf configure option should definitely be treated like
a variant of netcdf-4, the way Ed implemented it.

--Russ  

> I am getting very confused about pnetcdf support.
> This comment of yours
> >   2. For netCDF classic format or 64-bit offset format files that use the s
> econd nc_create_par() 
> >      call above, the parallel I/O uses pnetcdf aka parallel-netcdf, but pro
> vides the same API
> >      (and thus API parameter types) as in the serial netCDF API.  This rest
> ricts it to arrays of
> >      primitive types for data, rather than the structured MPI types of data
>  that pnetcdf can also 
> >      access, but makes it easier to convert a serial netCDF program to use 
> parallel I/O.
> makes me think that pnetcdf code should be in libsrc or as a separate dispatc
> h table.
> However the current code has all the pnetcdf stuff in libsrc4.
> When the --enable-pnetcdf is set for configure, should we treat this like a v
> ariant
> of netcdf-3 or netcdf-4?
> 
> =Dennis
> 
> 
> 
> Ticket Details
> ===================
> Ticket ID: UUS-259413
> Department: Support netCDF
> Priority: Normal
> Status: Open
> Link:  https://www.unidata.ucar.edu/esupport/staff/index.php?_m=tickets&_a=vi
> ewticket&ticketid=21396



Ticket Details
===================
Ticket ID: UUS-259413
Department: Support netCDF
Priority: Normal
Status: Open