[HACKERS] Optimizer Bug issue

2004-05-25 Thread Ismail Kizir
Tom, You say that default values for optimizer are efficient. But this is evident that it doesn'tfunction as expectedin my case. I have a database of 20 tables, ~1gb total size. My biggest table contains ~270,000 newspaper article from Turkish journals. I am actually working on "fulltext

Re: [HACKERS] Optimizer Bug issue

2004-05-25 Thread Greg Stark
Ismail Kizir [EMAIL PROTECTED] writes: I have a database of 20 tables, ~1gb total size. My biggest table contains ~270,000 newspaper article from Turkish journals. I am actually working on fulltext search program of my own. How much RAM does the machine have? Have you already executed the

Re: [HACKERS] Optimizer Bug issue

2004-05-25 Thread Tom Lane
Ismail Kizir [EMAIL PROTECTED] writes: I am sure default values are erronous. Perhaps they are --- for your database, on your hardware. Or perhaps the problem is somewhere else (we know that the costing of nestloop indexscan joins needs work, for instance). But in any case there's a reason why

Re: [HACKERS] Optimizer Bug issue

2004-05-25 Thread Christopher Kings-Lynne
I am actually working on fulltext search program of my own. No need, use contrib/tsearch2 Chris ---(end of broadcast)--- TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]