NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.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.

[netcdfgroup] aligning variables

I was wondering if there's a way in netcdf to ensure all variables
(record or non-record) start at an aligned offset?

I know with the v_align parameter to nc__enddef I can make sure the
netcdf (CDF-1) header is aligned to a certain size, and with the
chunksizehintp parameter to nc__create, data will be allocated in
'chunksize' pieces.  

Those are close to what I want, but what I'm looking for is a way to
force the beginning offset in file of every variable in my netcdf
dataset to a block-aligned value.  

Let's take GPFS for example: it uses megabyte blocks, and I've got
tons of space, so I want every variable, no matter how small, to begin
at a multiple of 1MB (1024*1024).   I don't care if there's slack
space between the end of one 16x16x16 integer array variable and the
next. Can I do this in netcdf?

Thanks
==rob

-- 
Rob Latham
Mathematics and Computer Science Division    A215 0178 EA2D B059 8CDF
Argonne National Lab, IL USA                 B29D F333 664A 4280 315B