Re: [netcdfgroup] nf90_var_par_access - independent or collective

Hi Karsten,

I also bumped into this problem (although I didn’t get any error message). If 
one of your dimension is unlimited then all parallel IO must be collective.

Cheers,

John

From: Karsten Bolding [mailto:karsten@xxxxxxxxxxxxxxxxxxxx]
Sent: 10 September 2014 11:26
To: netcdfgroup@xxxxxxxxxxxxxxxx
Subject: [netcdfgroup] nf90_var_par_access - independent or collective

Hello

Compiling like:
mpif90 simple_xy_par_wr2.f90 -I/opt/netcdff-4.4.1-gfortran-4.8-par/include/ 
/opt/netcdff-4.4.1-gfortran-4.8-par/lib/libnetcdff.a  
/opt/netcdf-4.3.3-rc1-gcc-4.8-par/lib/libnetcdf.a  
/opt/hdf5-1.8.13-gcc-4.8-par/lib/libhdf5_hl.a 
/opt/hdf5-1.8.13-gcc-4.8-par/lib/libhdf5.a -ldl -lm -lz -lcurl

I can run:

mpiexec -n 10 ./a.out

with collective it works but with independent I get:

 NetCDF: Attempt to extend dataset during NC_INDEPENDENT I/O operation. Use 
nc_va
 NetCDF: Attempt to extend dataset during NC_INDEPENDENT I/O operation. Use 
nc_va
 NetCDF: Attempt to extend dataset during NC_INDEPENDENT I/O operation. Use 
nc_va
STOP 2
 NetCDF: Attempt to extend dataset during NC_INDEPENDENT I/O operation. Use 
nc_va

any ideas?

Karsten



________________________________
This message (and any attachments) is for the recipient only. NERC is subject 
to the Freedom of Information Act 2000 and the contents of this email and any 
reply you make may be disclosed by NERC unless it is exempt from release under 
the Act. Any material supplied to NERC may be stored in an electronic records 
management system.
  • 2014 messages navigation, sorted by:
    1. Thread
    2. Subject
    3. Author
    4. Date
    5. ↑ Table Of Contents
  • Search the netcdfgroup archives: