From: | Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us> |
---|---|
To: | Dustin Sallings <dustin(at)spy(dot)net> |
Cc: | The Hermit Hacker <scrappy(at)hub(dot)org>, pgsql-general(at)hub(dot)org |
Subject: | Re: [GENERAL] Benchmarks |
Date: | 2000-01-06 18:14:19 |
Message-ID: | 200001061814.NAA15845@candle.pha.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
> Untrue, vacuum is *extremely* important for updating statistics.
> If you have a lot of data in a table, and you have never vacuumed, you
> might as well not have any indices. It'd be nice if you could seperate
> the stat update from the storage reclaim. Actually, it'd be nice if you
> could reuse storage, so that an actual vacuum wouldn't be necessary unless
> you just wanted to free up disk space you might end up using again anyway.
>
> The vacuum also doesn't seem to be very efficient. In one of my
> databases, a vacuum could take in excess of 24 hours, while I've written a
> small SQL script that does a select rename and a insert into select from
> that will do the same job in about ten minutes. This is a database that
> cannot lock for more than a few minutes.
This is serious. Why would an INSERT / RENAME be so much faster. Are
we that bad with VACUUM?
--
Bruce Momjian | http://www.op.net/~candle
maillist(at)candle(dot)pha(dot)pa(dot)us | (610) 853-3000
+ If your life is a hard drive, | 830 Blythe Avenue
+ Christ can be your backup. | Drexel Hill, Pennsylvania 19026
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2000-01-06 19:03:55 | Re: [BUGS] problem creating index in 6,5,3 |
Previous Message | Dustin Sallings | 2000-01-06 18:04:19 | Re: [GENERAL] Benchmarks |