| From: | Francisco Reyes <lists(at)natserv(dot)com> |
|---|---|
| To: | Gurupartap Davis <partap(at)yahoo(dot)com> |
| Cc: | PostgreSQL-General <pgsql-general(at)postgresql(dot)org> |
| Subject: | Re: Optimize for insertions? |
| Date: | 2001-11-27 15:38:07 |
| Message-ID: | 20011127103232.S95906-100000@zoraida.natserv.net |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
On Mon, 26 Nov 2001, Gurupartap Davis wrote:
> I'm worried about creating indexes after all the data is in there,
> though...I'm going to need 4-5 indexes. Isn't that going to take a hella
> long time? How are indexes created anyhow? I would assume a linear search
> through every record in the table...would vacuum help at all? how bout
> vacuum analyze?
I am somewhat of a newbie, but I was told that 'copy' doesn't use the
indexes so you could create your indexes and when you are all done with
the copy commands then run a vacuum analyze.
I do bulk uploads daily from a Foxpro system, been 7million my biggest
table, and I have found that a plain vacuum is helpfull after each bulk upload.
When I am done with all my bulk uploads, 3 files, then I do one vacuum analyze.
In theory you could just do a single vacuum analyze when you are all done,
but for me a vacuum after each upload seems to help. In particular I am
still trying to decide whether to use truncate or delete's. When I use
delete old the old records seems to still "be around" so the vacuum gets
rid of them. The main advantage of truncate is that it is a fast way to
empty a table, it's main drawback is that it can not be rolled back.
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Francisco Reyes | 2001-11-27 15:44:58 | Re: Timing a query |
| Previous Message | William WAISSE | 2001-11-27 15:37:39 | Re: newbie needs help with postgres internal tables |