Re: TableSpace Design issues on Postgres 8.0 beta 5

From: Rosny
Subject: Re: TableSpace Design issues on Postgres 8.0 beta 5
Date: ,
Msg-id: cp3nqc$27fl$1@news.hub.org
(view: Whole thread, Raw)
In response to: TableSpace Design issues on Postgres 8.0 beta 5  ("Rosny")
List: pgsql-performance

Thanks for the response.

I just start to get a feel of where TableSpace will be used. You are right I
do not have 7000 logical filesystems.
I am assuming  using TableSpace as organization of files in folders in
Windows 2003 Environment. So, each
TableSpace will represent one folder(directory) in a drive. But this doesnot
work to well for my design.

Since pgAdmin III GUI take 2 hours just to load approximately 14000 tables.
I am not using TableSpace approach anymore. I am using multiple database
approach.

Anyway, thanks for the response. Today testing by taking several option to
the limit. I kind of having some ideas
for future scability.

Rosny


"Tom Lane" <> wrote in message
news:...
> "Rosny" <> writes:
> > B. For each department create tablespace. (Which means, if there
> >  are 7000 departments, there will be 7000 tablespace each contains
> >  20 tables).
>
> If your system has seven thousand separate logical filesystems attached
> to it, there might be some value in having seven thousand tablespaces.
> But I will bet a great deal that it does not and there isn't.
>
> regards, tom lane
>
> ---------------------------(end of broadcast)---------------------------
> TIP 4: Don't 'kill -9' the postmaster
>




pgsql-performance by date:

From: Sven Willenberger
Date:
Subject: Using LIMIT changes index used by planner
From: Andrew McMillan
Date:
Subject: Re: Using LIMIT changes index used by planner