Depending on your granularity, storing PDLs as either blobs or as  
files whose names are stored in the DB could work for you.

Also check out PDL::DiskCache, which is a tied array object for quasi- 
transparently keeping PDLs on disk.  (DiskCache uses FITS files by  
default but can use Storable with minimal effort).



On Jun 22, 2010, at 2:49 PM, P Kishor wrote:

> On Tue, Jun 22, 2010 at 3:32 PM, David Mertens <[email protected] 
> > wrote:
>>
>>
>> On Tue, Jun 22, 2010 at 3:10 PM, P Kishor <[email protected]>  
>> wrote:
>>>
>>> On Tue, Jun 22, 2010 at 3:01 PM, Craig DeForest
>>> <[email protected]> wrote:
>>>> Looks like you're storing integer data in a double-precision
>>>> floating-point
>>>> PDL.  (default data type is 'double').  Try using type  
>>>> declarations in
>>>> your
>>>> constructors -- e.g. "ones(short, $x)->dummy * pdl(short, @dm)"
>>>>
>>>
>>> ahhh...
>>>
>>> -rw-r--r--   1 punkish  punkish    1289616 Jun 22 14:42 bin.dat
>>> -rw-r--r--   1 punkish  punkish    1280000 Jun 22 15:07 pdl.dat
>>> -rw-r--r--   1 punkish  punkish         14 Jun 22 15:07 pdl.dat.hdr
>>>
>>>
>>> So much better. So, a modest technical question -- is a piddle,
>>> internally, just like a packed value with some magic dust  
>>> sprinkled on
>>> it?
>>
>> Yes. Actually quite a bit of magic dust, and some lint. You  
>> shouldn't go
>> around creating thousands of 5-element piddles because they would  
>> probably
>> be more efficiently stored as perl arrays. However, if you cat'd  
>> those
>> piddles together, then you'd be in business.
>>
>
>
> Thanks David. For now, I am trying different things out, and,
> hopefully, learning how to "PDL walk" in the process. Actually, even
> my little-trainer-wheels test code is actually creating a "few hundred
> X few hundred piddle" of 6 element piddles. Since I potentially need
> to use PDL to work on those data once I take them out and off of the
> disk, I am guessing PDL will be way more efficient than working with
> array of arrays and Storable. RIght now I am researching potential
> alternatives to db storage as well as piddle+db storage... just trying
> out different approaches.
>
>
>> In terms of storage, however, the stuff stored to the disk is just  
>> the data.
>>
>> David
>>
>> --
>> Sent via my carrier pigeon.
>>
>
>
>
> -- 
> Puneet Kishor http://www.punkish.org
> Carbon Model http://carbonmodel.org
> Charter Member, Open Source Geospatial Foundation http://www.osgeo.org
> Science Commons Fellow, http://sciencecommons.org/about/whoweare/ 
> kishor
> Nelson Institute, UW-Madison http://www.nelson.wisc.edu
> -----------------------------------------------------------------------
> Assertions are politics; backing up assertions with evidence is  
> science
> = 
> ======================================================================
>
> _______________________________________________
> Perldl mailing list
> [email protected]
> http://mailman.jach.hawaii.edu/mailman/listinfo/perldl
>


_______________________________________________
Perldl mailing list
[email protected]
http://mailman.jach.hawaii.edu/mailman/listinfo/perldl

Reply via email to