From: | Andrew Sullivan <ajs(at)crankycanuck(dot)ca> |
---|---|
To: | pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: Tablespaces |
Date: | 2004-03-01 14:31:32 |
Message-ID: | 20040301143132.GB8345@phlogiston.dyndns.org |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers pgsql-hackers-win32 |
On Thu, Feb 26, 2004 at 05:28:41PM -0500, Alex J. Avriette wrote:
> The only reason I mentioned it to begin with was the recommendation of
> directio for databases in the Sun Blueprint, _Tuning Databases on the
> Solaris Platform_ (and being a Solaris geek, I asked, but apparently
> nobody else is worried enough about performance or not using Solaris
> enough to care).
That recommendation itself is a few years old. While it may still be
true that directio is still fastest for Oracle on Solaris, I'd sure
like to see some recent evidence. I've a funny feeling that this is
an old rule of thumb which is now true in the sense that everyone
believes it, but maybe not in the sense that a test would reveal it
to be a sensible rule.
> like to see some of the features of Oracle and DB2 available to users
> of postgres. Some of these features are raw disk, tablespaces, and
> replication. We're getting there, and making really terrific progress
I don't think we want features for their own sake, though, and I'm
not convinced that raw filesystems are actually useful. Course, it's
not my itch, and PostgreSQL _is_ free software.
A
--
Andrew Sullivan
From | Date | Subject | |
---|---|---|---|
Next Message | Andrew Sullivan | 2004-03-01 15:09:04 | Re: [HACKERS] Collaboration Tool Proposal |
Previous Message | Martin Marques | 2004-03-01 14:21:52 | Re: lib for clients |
From | Date | Subject | |
---|---|---|---|
Next Message | Andrew Sullivan | 2004-03-01 15:09:04 | Re: [HACKERS] Collaboration Tool Proposal |
Previous Message | Andrew Dunstan | 2004-03-01 10:19:21 | Re: [HACKERS] Collaboration Tool Proposal -- Summary to date |