Re: [PERFORM] Very big insert/join performance problem (bacula)

2009-07-13 Thread Marc Cousin
> > While this is not your questions, I still noticed you seem to be on 8.3 - > it might be a bit faster to use GROUP BY instead of DISTINCT. It didn't do a big difference, I already tried that before for this query. Anyway, as you said, it's not the query having problems :) > Your effective_cac

[PERFORM] Odd performance / query plan with bitmasked field as opposed to equality

2009-07-13 Thread Frank Joerdens
I can't figure what is going on below; first of all, this count which returns 1.5 million from a ~2 million row table: woome=# explain analyze SELECT COUNT(*) FROM "webapp_person" WHERE "webapp_person"."permissionflags" = B''::"bit";

Re: [PERFORM] embedded sql regression from 8.2.4 to 8.3.7

2009-07-13 Thread Haszlakiewicz, Eric
>-Original Message- >From: Albe Laurenz [mailto:laurenz.a...@wien.gv.at] >Eric Haszlakiewicz wrote: >> const char *SQL_text = "select * from foo"; (not always >the same query) >> exec sql prepare s_1ab from :SQL_text; < [*1] >> exec sql declare c_1ab cursor for s_1ab; >

Re: [PERFORM] embedded sql regression from 8.2.4 to 8.3.7

2009-07-13 Thread Haszlakiewicz, Eric
>-Original Message- >From: Albe Laurenz [mailto:laurenz.a...@wien.gv.at] >> >> const char *SQL_text = "select * from foo"; (not always >the same query) >> exec sql prepare s_1ab from :SQL_text; < [*1] >> exec sql declare c_1ab cursor for s_1ab; >> exec sql open c_1

Re: [PERFORM] Very big insert/join performance problem (bacula)

2009-07-13 Thread Andres Freund
Hi Marc, I don't have really extensive comments, but I found two small things... On Monday 13 July 2009 15:40:18 Marc Cousin wrote: > I'm trying to solve big performance issues with PostgreSQL + bacula while > inserting very big sets of records. > > I'm sorry, this email will be a long one, as I'

Re: [PERFORM] Very big insert/join performance problem (bacula)

2009-07-13 Thread Marc Cousin
We regularly do all of dbcheck. This is our real configuration, there are really lots of servers and lots of files (500 million files backed up every month). But thanks for mentionning that. The thing is we're trying to improve bacula with postgresql in order to make it able to bear with this

Re: [PERFORM] Very big insert/join performance problem (bacula)

2009-07-13 Thread SystemManagement
Hi, just a remark, as the number of entries seems to be very high: Did you ever activate bacula's program dbcheck Option 16? Regards Reiner Marc Cousin schrieb: Hi, I'm trying to solve big performance issues with PostgreSQL + bacula while inserting very big sets of records. I'm sorry, thi

[PERFORM] Very big insert/join performance problem (bacula)

2009-07-13 Thread Marc Cousin
Hi, I'm trying to solve big performance issues with PostgreSQL + bacula while inserting very big sets of records. I'm sorry, this email will be a long one, as I've already spent quite a lot of time on the issue, I don't want to waste your time speculating on things I may already have done, and

Re: Res: [PERFORM] Cost performace question

2009-07-13 Thread Matthew Wakeling
On Fri, 10 Jul 2009, paulo matadr wrote: thanks i understand now,  i have other dought Index Scan using xix3_movimento_roteiro_empresa on movimento_roteiro_empresa movimentor0_  (cost=0.00..73386.80 rows=5126 width=451) what's this mean on red color above? Please don't assume that everyone rec