Re: Status of server side Large Object support? - Mailing list pgsql-hackers

From Bruce Momjian
Subject Re: Status of server side Large Object support?
Date
Msg-id 200412130307.iBD379w04484@candle.pha.pa.us
Whole thread Raw
In response to Re: Status of server side Large Object support?  (Hannu Krosing <hannu@tm.ee>)
List pgsql-hackers
Hannu Krosing wrote:
> On E, 2004-11-29 at 02:22, David Garamond wrote:
> > Joe Conway wrote:
> > > Not if the column is storage type EXTERNAL. See a past discussion here:
> > > http://archives.postgresql.org/pgsql-general/2003-07/msg01447.php
> > 
> > what is the reasoning behind this syntax?
> > 
> >   ALTER TABLE [ ONLY ] table [ * ]
> >   ALTER [ COLUMN ] column SET STORAGE
> >   { PLAIN | EXTERNAL | EXTENDED | MAIN }
> > 
> > I find it nonintuitive and hard to remember. Perhaps something like this 
> > is better (I know, it's probably too late):
> > 
> >   ALTER [ COLUMN ] column SET STORAGE { INLINE | EXTERNAL }
> >   ALTER [ COLUMN ] column SET COMPRESSION { YES | NO }
> 
> It wold also be beneficial if the threshold size of moving the column to
> TOAST (either COMPRESS or EXTERNAL) could be set on a per-column basis
> 
> This is a design decision on the same lavel as the others

The threshhold is currently per-row, with longer rows being toasted
first. I can't see having a per-column specification being a big win.

--  Bruce Momjian                        |  http://candle.pha.pa.us pgman@candle.pha.pa.us               |  (610)
359-1001+  If your life is a hard drive,     |  13 Roberts Road +  Christ can be your backup.        |  Newtown Square,
Pennsylvania19073
 


pgsql-hackers by date:

Previous
From: Neil Conway
Date:
Subject: Re: [Testperf-general] BufferSync and bgwriter
Next
From: Mark Kirkwood
Date:
Subject: Re: [Testperf-general] BufferSync and bgwriter