[postgis-users] upgrade benefits re. high load

[Paul Ramsey]

It was indeed a backhanded reference to lwgeom, whose primary benefit 
is to do everything hwgeom does, but in much less space. So, if you 
have a really huge database, the lwgeom version of it will be smaller, 
as will the indexes. If your problem is one of query load, assuming you 
have a large amount of memory in your current production database, 
lwgeom probably won't make a large difference, because all the 
important stuff relating to your hwgeom database is probably cached in 
memory anyways.

Have you done any tuning to your pgsql installation at all? Possibly 
some tuning of your database server would be sufficient to make you get 
better performance, without the need to move on to lwgeom. Really, if 
you don't have > 1million tuples in your database, lwgeom is unlikely 
to make much difference to you one way or another.

On Thursday, September 30, 2004, at 07:19 AM, Purvis, Charlton wrote:

> Anyway, if one of you czars out there wouldn't mind letting me know 
> what was meant by high load improvement, I'd be thankful.  I'm 
> woefully afraid of having to drop the db's to upgrade, so it would 
> have to be a bright light at the end of that tunnel to push me in that 
> direction.

      Paul Ramsey
      Refractions Research
      Email: pramsey at refractions.net
      Phone: (250) 885-0632