From: | "Brian Maguire" <bmaguire(at)vantage(dot)com> |
---|---|
To: | <lnd(at)hnit(dot)is> |
Cc: | <pgsql-general(at)postgresql(dot)org> |
Subject: | Re: tablespaces a priority for 7.5? |
Date: | 2004-01-22 02:03:15 |
Message-ID: | 203C7FC3FF2D7A4588CE0429A87F3C9A0AC567@vt-pe2550-001.vantage.vantage.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
I agree that RAID provides similar performance benifits especially with striping io benifits, however it is powerful and ideal to have both options. For example you may have a set of tables that are read-only for reporting and another set mostly write only. You could have they resting on different raid configurations ideal for each situtation.
I also agree there are several admin benifits in the areas of backup. You can also more easily create different frequency/schedules of backup for certain critical tables to a different schedule than other less important tables. The flexibility of easily growing your database beyond the current disk because of size limitations can be a life savior.
brian
-----Original Message-----
From: lnd(at)hnit(dot)is [mailto:lnd(at)hnit(dot)is]
Sent: Wed 1/21/2004 12:22 PM
To: Brian Maguire
Cc:
Subject: RE: [GENERAL] tablespaces a priority for 7.5?
In RAID era tablespaces are not such important regarding performance.
But for backup/restore - the ability to backup/restore selected tablespaces
while leaving other tablespaces is a big thing.
The whole point here is: it is assumed that backup/restore of tablespaces can
hapen quite quickly and as simple as to copy tablespace files from one
location to another(even while database is on - WAL can be used to handle
this) - this is compared to dump.
For example, index, tempoarary data tablespaces can be lost - not a big deal.
Undo(rollback) tablespaces - in a way can be lost as well.
While system data tablespace (table structure, stored procedures, etc) - at
no cost should be lost.
The same way application can be devided in "critical" and "not critical"
tablespaces and their backups maintained accordingly. For example, it may not
be a big deal to lose year 1996 tables while year 2004 tables should be
online.
Laimis
> -----Original Message-----
> From: pgsql-general-owner(at)postgresql(dot)org
> [mailto:pgsql-general-owner(at)postgresql(dot)org] On Behalf Of Brian Maguire
> Sent: 21. janúar 2004 16:06
> To: pgsql-general(at)postgresql(dot)org
> Subject: [GENERAL] tablespaces a priority for 7.5?
>
>
> Is support for tablespaces a priority feature for 7.5? I
> believe there has been significant development in this area
> and it seems that postgres' file structure opens it up nicely
> to support it. What are the chances this will be completed?
>
> In my opinion, it really is a critical feature to support and
> administer enterprise databases. All the major databases
> currently support this and it is a compelling enough reason
> drive big users from away from using postgres for their
> enterprise/large databases. It really is a database
> administrator's feature.
>
>
> Brian
>
>
> ---------------------------(end of
> broadcast)---------------------------
> TIP 6: Have you searched our list archives?
>
http://archives.postgresql.org
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2004-01-22 03:12:29 | Re: tablespaces a priority for 7.5? |
Previous Message | Richard Huxton | 2004-01-22 01:54:02 | Re: postgresql + apache under heavy load |