Re: Memory issue: Does linint2_Wrap hold on to anything in memory? Does dble2flt conversion allocate more memory than simple float array?

From: Dennis Shea <shea_at_nyahnyahspammersnyahnyah>
Date: Thu Oct 22 2009 - 13:32:13 MDT

linint2_Wrap should have *trivial* memory overhead.

How much memory do you have available?

When you have
     dvar = dble2flt(fdata->$var$)

How big is
     x = fdata->$var$
     printVarSummary(x)

The dble2flt creates a new array that is half the (byte) size of the
double array [ 8 bytes for double and 4 bytes for float].

    dvar = dble2flt (x )

I do not know how NCL allocates memory for this type of tranfer.

D

Bridget Thrasher wrote:
> I am being confronted with the dreaded malloc error when performing
> the following steps:
>
> ...
> factors = fdata->factors ; this variable is a float
> factors_int = linint2_Wrap(lons,lats,factors,True,obs_lons,obs_lats,0)
> delete(factors)
> bcsd->$var$ = factors_int
> delete(factors_int)
> dvar = dble2flt(fdata->$var$)
> dvar_int = linint2_Wrap(lons,lats,dvar,True,obs_lons,obs_lats,0)
> ...
>
> This runs fine until the last statement, at which point I get this:
>
> ncl(26119) malloc: *** mmap(size=1866240000) failed (error code=12)
> *** error: can't allocate region
> *** set a breakpoint in malloc_error_break to debug
> fatal:linint2: Unable to allocate memory for output array
> fatal:Execute: Error occurred at or near line 3784 in file
> $NCARG_ROOT/lib/ncarg/nclscripts/csm/contributed.ncl
>
> Since I'm deleting the previous arrays, and the result of the
> interpolation should be the same size, maybe there is some
> intermediate step of the first interpolation is still in memory? Or
> does dble2flt allocate an array of larger memory size than that
> allocated when reading in a float variable of the same dimensions? Or
> is there something else I'm overlooking?
>
> Thanks for your time,
> Bridget
>
>
> --
> Bridget Thrasher, PhD
> Postdoctoral Researcher
> Climate Central
> www.climatecentral.org <http://www.climatecentral.org>
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> ncl-talk mailing list
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>

-- 
======================================================
Dennis J. Shea                  tel: 303-497-1361    |
P.O. Box 3000                   fax: 303-497-1333    |
Climate Analysis Section                             |
Climate & Global Dynamics Div.                       |
National Center for Atmospheric Research             |
Boulder, CO  80307                                   |
USA                        email: shea 'at' ucar.edu |
======================================================
_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Thu Oct 22 13:32:11 2009

This archive was generated by hypermail 2.1.8 : Fri Oct 23 2009 - 11:36:30 MDT