From: | Chris Browne <cbbrowne(at)acm(dot)org> |
---|---|
To: | pgsql-general(at)postgresql(dot)org |
Subject: | Re: VACUUM Question |
Date: | 2006-01-26 21:14:45 |
Message-ID: | 603bjan12y.fsf@dba2.int.libertyrms.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general pgsql-hackers |
matthew(at)zeut(dot)net ("Matthew T. O'Connor") writes:
> If you really are just inserting, and never updating or deleting,
> then you will never need to vacuum the table, rather you will just
> need to ANALYSE the table. If you use autovacuum that is exactly
> what it will do.
"Never" is a pretty long time...
You need a VACUUM every 2^31 transactions, but since there needs to be
such a vacuum for the whole database, that one will do...
> As for Reindex, I'm not entirely sure, I don't think you would benefit
> from reindex because you aren't updating or deleting. Can anyone comment
> on this? Is is possibile that a table with lots of inserts resulting in
> lots of page splits etc could ever benifit form REINDEX?
I could imagine a CLUSTER doing some good, and if that's the case,
REINDEX could have some favorable results. But you'd better have a
real specific model as to why that would be...
--
let name="cbbrowne" and tld="ntlug.org" in name ^ "@" ^ tld;;
http://cbbrowne.com/info/spreadsheets.html
Oh, boy, virtual memory! Now I'm gonna make myself a really *big*
RAMdisk!
From | Date | Subject | |
---|---|---|---|
Next Message | Eric B. Ridge | 2006-01-26 21:19:34 | "xmin" system column |
Previous Message | Tino Wildenhain | 2006-01-26 21:09:34 | Re: Arrays |
From | Date | Subject | |
---|---|---|---|
Next Message | Bruce Momjian | 2006-01-26 21:31:09 | Re: -X flag in pg_dump |
Previous Message | Peter Eisentraut | 2006-01-26 21:12:01 | Re: Cleaning up the INET/CIDR mess |