NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.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NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.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.
Hi Ted The parallel I/O of NetCDF 4.6.0 has been revised quite a lot. Constants NC_MPIIO, NC_PNETCDF, and NC_MPIPOSIX are deprecated in 4.6.0. Because parallel I/O in NetCDF must be done by calling nc_create_par or nc_open_par, these 3 constants become redundant. When calling nc_create_par, the format of file to be created is determined by flags NC_NETCDF4, NC_CLASSIC_MODEL, NC_64BIT_OFFSET, and NC_64BIT_DATA. The former two are for HDF5-based files and latter two are classic-based. For nc_open_par, NetCDF checks the file signature which decides either to call the HDF5 driver or PnetCDF drivers underneath to carry out the parallel I/O. I guess the next release of NetCDF-Fortran will be updated to reflect the above changes in C. FYI. command-line "nc-config --version” and "nf-config --version” print the library version number. So are APIs nf90_inq_libvers and nf_inq_libvers. Wei-keng > On Jan 31, 2019, at 6:27 PM, Ted Mansell <ted.mansell@xxxxxxxx> wrote: > > Hi, > > I have been using older versions of netcdf for a long time (4.3.1.1) and just > recently tried out 4.6.0. I sometimes use pnetcdf for writing, in which case > I used the convenient netcdf-4 interface to define the files like so: > > cmode = ior(ior(NF90_PNETCDF,NF90_MPIIO),nf90_64bit_offset) > status = nf90_create(filename(ibeg:iend), cmode, ncid, comm = my_comm, > info = my_info) > > With 4.6.0, the NF90_PNETCDF causes an error, but it works if I remove that > flag: > cmode = ior(NF90_MPIIO,nf90_64bit_offset) > > But that doesn't work with the older netcdf. Is there a way to check which > version of netcdf I'm compiling against so that I can adapt the mode > appropriately? I've been testing out ideas with the f90tst_parallel.f90, but > can't seem to get both versions to work with the same code. I'm too lazy to > write a separate code to create and define my file with pure pnetcdf.... > > Thanks for any hints. > > -- Ted > > __________________________________________________________ > | Edward Mansell <ted.mansell@xxxxxxxx> > | National Severe Storms Laboratory > |-------------------------------------------------------------- > | "The contents of this message are mine personally and > | do not reflect any position of the U.S. Government or NOAA." > |-------------------------------------------------------------- > > _______________________________________________ > NOTE: All exchanges posted to Unidata maintained email lists are > recorded in the Unidata inquiry tracking system and made publicly > available through the web. Users who post to any of the lists we > maintain are reminded to remove any personal information that they > do not want to be made public. > > > netcdfgroup mailing list > netcdfgroup@xxxxxxxxxxxxxxxx > For list information or to unsubscribe, visit: > http://www.unidata.ucar.edu/mailing_lists/
netcdfgroup archives: