[Met_help] [rt.rap.ucar.edu #75780] History for Re: Verification of GFS data
John Halley Gotway via RT
met_help at ucar.edu
Mon Apr 18 12:18:05 MDT 2016
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
Thanks! I tried using the mask section, but I still get the same message.
It seems to be failing here, just after reading the config file. I think
this happens before the mask is set.
DEBUG 4: Met2dDataFileFactory::new_met_2d_data_file() -> created new
Met2dDataFile object of type "Fi
leType_Gb2".
ERROR :
ERROR : out of memory! Exiting!
ERROR :
The gfs file is in grib2 Format. Is that a problem? Our local
version of MET (Version 5.1) was compiled with grib2 Support.
Is it possible to run the MET Tools in parallel?
Thanks
Carlos
On Tuesday, April 5, 2016, Randy Bullock via RT <met_help at ucar.edu> wrote:
> Hello!
>
> Sorry for the delay in replying ... most of our people are out for spring
> break.
>
> Yes, there is a way to select a particular region. You can specify a
> masking region in the point_stat config file. Take a look at section 7.3.2
> in the MET version 5.1 user's manual. Or, if you're using an earlier
> version of MET, just look in the section that describes the point_stat
> config file. The "mask" entry in the config file is the one you need.
>
> Hope this helps.
>
> Randy
>
> On Tue, Apr 5, 2016 at 2:05 AM, Carlos Peralta via RT <met_help at ucar.edu>
> wrote:
>
> >
> > Tue Apr 05 02:05:29 2016: Request 75759 was acted upon.
> > Transaction: Ticket created by carlos9917 at gmail.com
> > Queue: met_help
> > Subject: Verification of GFS data
> > Owner: Nobody
> > Requestors: carlos9917 at gmail.com
> > Status: new
> > Ticket <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=75759 >
> >
> >
> > Hello,
> >
> > I am trying to use point_stat to verify temperature and wind over
> > selected stations using GFS data I downloaded from this Location:
> > http://www.nco.ncep.noaa.gov/pmb/products/gfs/
> >
> > However, I keep getting the message:
> >
> > ERROR :
> > ERROR : out of memory! Exiting!
> > ERROR :
> >
> > I guess the GFS file is too big? Each GFS file is about 217 MB, and I
> have
> > used WRF files way larger than that before and Point_stat worked fine.
> > Is there a way to select only a particular region?
> >
> > Thanks heaps in advance for any help on this issue.
> >
> > Carlos
> >
> >
>
>
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: Re: Verification of GFS data
From: Randy Bullock
Time: Wed Apr 06 09:05:31 2016
Hi again -
If you compiled MET with the grib2 support option, then grib2 files
should
not be a problem. If that were the source of your memory issue, you
wouldn't get an out-of-memory error message ... the message would be
something about an unrecognized file format.
As to your question about parallelizing MET, the MET code wasn't
written in
such a way as to facilitate parallelization. That's maybe something
our
group should look at in the future.
Concerning the memory problem, one other thing you might check is the
size
of the grid. If the grid has very many points, that could conceivably
cause the computer to run out of memory. If that's the case, perhaps
you
could regrid your data to a smaller grid.
Failing that, I can only suggest trying to run MET on a different
computer
... one with more memory.
Sorry I couldn't fix this for you.
Randy
On Wed, Apr 6, 2016 at 4:57 AM, Carlos Peralta via RT
<met_help at ucar.edu>
wrote:
>
> Wed Apr 06 04:57:15 2016: Request 75780 was acted upon.
> Transaction: Ticket created by carlos9917 at gmail.com
> Queue: met_help
> Subject: Re: Verification of GFS data
> Owner: Nobody
> Requestors: carlos9917 at gmail.com
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=75780 >
>
>
> Thanks! I tried using the mask section, but I still get the same
message.
> It seems to be failing here, just after reading the config file. I
think
> this happens before the mask is set.
>
> DEBUG 4: Met2dDataFileFactory::new_met_2d_data_file() -> created new
> Met2dDataFile object of type "Fi
> leType_Gb2".
> ERROR :
> ERROR : out of memory! Exiting!
> ERROR :
>
> The gfs file is in grib2 Format. Is that a problem? Our local
> version of MET (Version 5.1) was compiled with grib2 Support.
>
> Is it possible to run the MET Tools in parallel?
>
> Thanks
>
> Carlos
>
>
>
> On Tuesday, April 5, 2016, Randy Bullock via RT <met_help at ucar.edu>
wrote:
>
> > Hello!
> >
> > Sorry for the delay in replying ... most of our people are out for
spring
> > break.
> >
> > Yes, there is a way to select a particular region. You can
specify a
> > masking region in the point_stat config file. Take a look at
section
> 7.3.2
> > in the MET version 5.1 user's manual. Or, if you're using an
earlier
> > version of MET, just look in the section that describes the
point_stat
> > config file. The "mask" entry in the config file is the one you
need.
> >
> > Hope this helps.
> >
> > Randy
> >
> > On Tue, Apr 5, 2016 at 2:05 AM, Carlos Peralta via RT
<met_help at ucar.edu
> >
> > wrote:
> >
> > >
> > > Tue Apr 05 02:05:29 2016: Request 75759 was acted upon.
> > > Transaction: Ticket created by carlos9917 at gmail.com
> > > Queue: met_help
> > > Subject: Verification of GFS data
> > > Owner: Nobody
> > > Requestors: carlos9917 at gmail.com
> > > Status: new
> > > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=75759
> >
> > >
> > >
> > > Hello,
> > >
> > > I am trying to use point_stat to verify temperature and wind
over
> > > selected stations using GFS data I downloaded from this
Location:
> > > http://www.nco.ncep.noaa.gov/pmb/products/gfs/
> > >
> > > However, I keep getting the message:
> > >
> > > ERROR :
> > > ERROR : out of memory! Exiting!
> > > ERROR :
> > >
> > > I guess the GFS file is too big? Each GFS file is about 217 MB,
and I
> > have
> > > used WRF files way larger than that before and Point_stat worked
fine.
> > > Is there a way to select only a particular region?
> > >
> > > Thanks heaps in advance for any help on this issue.
> > >
> > > Carlos
> > >
> > >
> >
> >
>
>
------------------------------------------------
Subject: Re: Verification of GFS data
From: Carlos Peralta
Time: Thu Apr 07 04:35:01 2016
Hello,
this problem was fixed by our local administrator.
It turns out there was a flag related to grib2
that was missing during the MET compilation.
All is working fine now with the recompiled MET and
I don't get any more "out of Memory" Problems :-)
.
Thanks again for your help.
Carlos
On Wed, Apr 6, 2016 at 5:05 PM, Randy Bullock via RT
<met_help at ucar.edu>
wrote:
> Hi again -
>
> If you compiled MET with the grib2 support option, then grib2 files
should
> not be a problem. If that were the source of your memory issue, you
> wouldn't get an out-of-memory error message ... the message would be
> something about an unrecognized file format.
>
> As to your question about parallelizing MET, the MET code wasn't
written in
> such a way as to facilitate parallelization. That's maybe something
our
> group should look at in the future.
>
> Concerning the memory problem, one other thing you might check is
the size
> of the grid. If the grid has very many points, that could
conceivably
> cause the computer to run out of memory. If that's the case,
perhaps you
> could regrid your data to a smaller grid.
>
> Failing that, I can only suggest trying to run MET on a different
computer
> ... one with more memory.
>
> Sorry I couldn't fix this for you.
>
> Randy
>
> On Wed, Apr 6, 2016 at 4:57 AM, Carlos Peralta via RT
<met_help at ucar.edu>
> wrote:
>
> >
> > Wed Apr 06 04:57:15 2016: Request 75780 was acted upon.
> > Transaction: Ticket created by carlos9917 at gmail.com
> > Queue: met_help
> > Subject: Re: Verification of GFS data
> > Owner: Nobody
> > Requestors: carlos9917 at gmail.com
> > Status: new
> > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=75780 >
> >
> >
> > Thanks! I tried using the mask section, but I still get the same
message.
> > It seems to be failing here, just after reading the config file. I
think
> > this happens before the mask is set.
> >
> > DEBUG 4: Met2dDataFileFactory::new_met_2d_data_file() -> created
new
> > Met2dDataFile object of type "Fi
> > leType_Gb2".
> > ERROR :
> > ERROR : out of memory! Exiting!
> > ERROR :
> >
> > The gfs file is in grib2 Format. Is that a problem? Our local
> > version of MET (Version 5.1) was compiled with grib2 Support.
> >
> > Is it possible to run the MET Tools in parallel?
> >
> > Thanks
> >
> > Carlos
> >
> >
> >
> > On Tuesday, April 5, 2016, Randy Bullock via RT
<met_help at ucar.edu>
> wrote:
> >
> > > Hello!
> > >
> > > Sorry for the delay in replying ... most of our people are out
for
> spring
> > > break.
> > >
> > > Yes, there is a way to select a particular region. You can
specify a
> > > masking region in the point_stat config file. Take a look at
section
> > 7.3.2
> > > in the MET version 5.1 user's manual. Or, if you're using an
earlier
> > > version of MET, just look in the section that describes the
point_stat
> > > config file. The "mask" entry in the config file is the one you
need.
> > >
> > > Hope this helps.
> > >
> > > Randy
> > >
> > > On Tue, Apr 5, 2016 at 2:05 AM, Carlos Peralta via RT <
> met_help at ucar.edu
> > >
> > > wrote:
> > >
> > > >
> > > > Tue Apr 05 02:05:29 2016: Request 75759 was acted upon.
> > > > Transaction: Ticket created by carlos9917 at gmail.com
> > > > Queue: met_help
> > > > Subject: Verification of GFS data
> > > > Owner: Nobody
> > > > Requestors: carlos9917 at gmail.com
> > > > Status: new
> > > > Ticket <URL:
> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=75759
> > >
> > > >
> > > >
> > > > Hello,
> > > >
> > > > I am trying to use point_stat to verify temperature and wind
over
> > > > selected stations using GFS data I downloaded from this
Location:
> > > > http://www.nco.ncep.noaa.gov/pmb/products/gfs/
> > > >
> > > > However, I keep getting the message:
> > > >
> > > > ERROR :
> > > > ERROR : out of memory! Exiting!
> > > > ERROR :
> > > >
> > > > I guess the GFS file is too big? Each GFS file is about 217
MB, and I
> > > have
> > > > used WRF files way larger than that before and Point_stat
worked
> fine.
> > > > Is there a way to select only a particular region?
> > > >
> > > > Thanks heaps in advance for any help on this issue.
> > > >
> > > > Carlos
> > > >
> > > >
> > >
> > >
> >
> >
>
>
------------------------------------------------
More information about the Met_help
mailing list