| From: | The Hermit Hacker <scrappy(at)hub(dot)org> |
|---|---|
| To: | bpalmer <bpalmer(at)crimelabs(dot)net> |
| Cc: | Alfred Perlstein <bright(at)wintelcom(dot)net>, pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: Why vacuum? |
| Date: | 2000-12-14 03:47:50 |
| Message-ID: | Pine.BSF.4.21.0012132345540.453-100000@thelab.hub.org |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Wed, 13 Dec 2000, bpalmer wrote:
> > Yes, postgresql requires vacuum quite often otherwise queries and
> > updates start taking ungodly amounts of time to complete. If you're
> > having problems because vacuum locks up your tables for too long
> > you might want to check out:
>
> But why? I don't know of other databases that need to be 'vacuum'ed. Do
> all others just do it internaly on a regular basis?
>
> What am I missing here?
PgSQL's storage manager is currently such that it doesn't overwrite
'deleted' records, but just keeps appending to the end of the table
... so, for instance, a client of ours whose table had 5 records in it
that are updated *alot* grew a table to 64Meg that only contains ~8k worth
of data ...
vacuum'ng cleans out the cruft and truncates the file ...
vadim, for v7.2, is planning on re-writing the storage manager to do
proper overwriting of deleted space, which will reduce the requirement for
vacuum to almost never ...
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tim Allen | 2000-12-14 03:58:25 | RE: Why vacuum? |
| Previous Message | bpalmer | 2000-12-14 03:45:05 | Re: Why vacuum? |