how to open a very large netcdf file using pgf90

[Wuhu Feng]

Dear Gradsusers,

I guess someone maybe help me to solve the following problem I met
recently.

Recently I am doing some simulations using an American mesoscale
model. It will be no problem to check the output using ncdump
command (attached). However, I encountered some problem to read the file
possible because it is too large(4972763028 bytes) when I tried to use a
Fortran program. It shows "758198573: heap block overflow; -heapz too
large Segmentation fault (core dumped)"

Here I use pgf90 (PGI4.0-Portland compilers v 4.0), netcdf-3.60 in Linux.

My PC is
"Linux portsdown 2.4.26-1-686-smp #1 SMP Mon Apr 19 14:12:30 CEST 2004
i686 unknown"

Below is how I compile and execute the program. Any suggestion will be
much appreciated.

Best regards,

Wuhu

portsdown:~> pgf90 read_wrf_nc.f
-L/nfs/env-fs-04_u22/fengwh/netcdf-3.60pl/lib -lnetcdf
-lm -I/nfs/env-fs-04_u22/fengwh/netcdf-3.60pl/include -Mfree -o
read_wrf_nc

portsdown:~> ./read_wrf_nc /nfs/env-fs-05_arc02/fengwh/WRFV2/run/wrfout_d01_2005-11-01_00:00:00
" INPUT FILE IS:
 /nfs/env-fs-05_arc02/fengwh/WRFV2/run/wrfout_d01_2005-11-01_00:00:00


 GLOBAL ATTRIBUTES:

TITLE                                    :  OUTPUT FROM WRF V2.1.1 MODEL
START_DATE                               : 2005-11-01_00:00:00
SIMULATION_START_DATE                    : 2005-11-01_00:00:00
WEST-EAST_GRID_DIMENSION                 :    91
SOUTH-NORTH_GRID_DIMENSION               :   121
BOTTOM-TOP_GRID_DIMENSION                :    31
758198573: heap block overflow; -heapz too large
Segmentation fault (core dumped)"

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: output_description.txt
Url: http://gradsusr.org/pipermail/gradsusr/attachments/20060222/47842f93/attachment.txt