From: | "Greg Sabino Mullane" <greg(at)turnstep(dot)com> |
---|---|
To: | pgsql-advocacy(at)postgresql(dot)org |
Subject: | Re: Restore times |
Date: | 2010-09-08 13:09:08 |
Message-ID: | c1fbb2c284d29a24ffb102251e1c76d4@biglumber.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-advocacy |
-----BEGIN PGP SIGNED MESSAGE-----
Hash: RIPEMD160
>> Their main database is in the 200-300GB size range and I just don't
>> have anything that size in production. Does anyone have stats on the
>> time taken to restore a database in that size range which they are
>> willing to share?
>
> Depends on the # of indexes. Ours take 2-3 hours on a really nice SAN
> storage.
Yeah, it *really* depends on the indexes. In my experience, most of
the time of restoring a database that size is the index and constraint
creation phase. The restore times are fairly linear however, so you should
be able to restore a smaller similar database on the same hardware and
extrapolate from there. Remember to boost maintenance_work_mem and
use the new pg_restore -j flag if you can.
As far as a time guess, I'd say 2-3 hours is probably on the low end of
most restores of that size, assuming a somewhat well-indexed system
consisting of a few score tables.
- --
Greg Sabino Mullane greg(at)turnstep(dot)com
End Point Corporation http://www.endpoint.com/
PGP Key: 0x14964AC8 201009080757
http://biglumber.com/x/web?pk=2529DF6AB8F79407E94445B4BC9B906714964AC8
-----BEGIN PGP SIGNATURE-----
iEYEAREDAAYFAkyHe0AACgkQvJuQZxSWSsh7jgCeLVDEpeXCfMEKR15Arm8gRkBw
izIAoPKMBPY/9r0vVZMNgLjyPtg3OKQ6
=bqlQ
-----END PGP SIGNATURE-----
From | Date | Subject | |
---|---|---|---|
Next Message | Joshua Kramer | 2010-09-10 00:59:48 | Article Review Requested |
Previous Message | Devrim GÜNDÜZ | 2010-09-08 09:19:06 | Re: Restore times |