Re: [PERFORM] Time to put theory to the test?

2011-04-25 Thread Mark Kirkwood
On 26/04/11 15:04, Rob Wultsch wrote: On Mon, Apr 25, 2011 at 12:30 PM, J Sisson wrote: machines, and virtually every MySQL machine has required data cleanup and table scans and tweaks to get it back to "production" status. Tip from someone that manages thousands of MySQL servers: Use InnoDB w

Re: [PERFORM] Performance

2011-04-25 Thread Robert Haas
On Apr 13, 2011, at 6:19 PM, Tomas Vondra wrote: > Yes, I've had some lectures on non-linear programming so I'm aware that > this won't work if the cost function has multiple extremes (walleys / > hills etc.) but I somehow suppose that's not the case of cost estimates. I think that supposition mi

Re: [PERFORM] Performance

2011-04-25 Thread Robert Haas
On Apr 14, 2011, at 2:49 AM, Claudio Freire wrote: > This particular factor is not about an abstract and opaque "Workload" > the server can't know about. It's about cache hit rate, and the server > can indeed measure that. The server can and does measure hit rates for the PG buffer pool, but to m

Re: [PERFORM] Time to put theory to the test?

2011-04-25 Thread Rob Wultsch
On Mon, Apr 25, 2011 at 12:30 PM, J Sisson wrote: > machines, and virtually every MySQL machine has required data cleanup > and table scans and tweaks to get it back to "production" status. Tip from someone that manages thousands of MySQL servers: Use InnoDB when using MySQL. Using a crash unsafe

Re: [PERFORM] Slow deleting tables with foreign keys

2011-04-25 Thread Robert Haas
On Wed, Mar 30, 2011 at 10:16 PM, Jeremy Palmer wrote: > Hi All, > > I'm trying to delete one row from a table and it's taking an extremely long > time. This parent table is referenced by other table's foreign keys, but the > particular row I'm trying to delete is not referenced any other rows i

Re: [PERFORM] %100 CPU on Windows Server 2003

2011-04-25 Thread Robert Haas
On Wed, Apr 20, 2011 at 10:55 AM, Allen Sooredoo < allen_soore...@carrefour.com> wrote: > Hi, > we are facing a performance issue on Postgres 8.4, the CPU reaches 100% > with less than 50 simultaneous users. > > We were thinking to migrate the HR system from Oracle to Postgres but now > that we ha

[PERFORM] Time to put theory to the test?

2011-04-25 Thread J Sisson
Not sure if this is the right list...but: Disclaimer: I realize this is comparing apples to oranges. I'm not trying to start a database flame-war. I just want to say thanks to the PostgreSQL developers who make my life easier. I manage thousands of databases (PostgreSQL, SQL Server, and MySQL)

Re: [PERFORM] big distinct clause vs. group by

2011-04-25 Thread Uwe Bartels
Hi Vitalii, this sounds promising, can you send me that? Best Regards, Uwe 2011/4/25 Віталій Тимчишин > > > 2011/4/23 Robert Haas > >> On Apr 18, 2011, at 1:13 PM, Uwe Bartels wrote: >> > Hi Robert, >> > >> > thanks for your answer. >> > the aggregate function I was talking about is the fun

Re: [PERFORM] Issue with partition elimination

2011-04-25 Thread Josh Berkus
On 4/21/11 6:26 PM, Paul Pierce wrote: > What was noticed, that the only case when Postgres is actually going to > execute > the query against the right partitions is query #1. > > Is that by design? Second query join, will also result in 3 days(3 > partitions) Partition elimination currentl

Re: [PERFORM] big distinct clause vs. group by

2011-04-25 Thread Віталій Тимчишин
2011/4/23 Robert Haas > On Apr 18, 2011, at 1:13 PM, Uwe Bartels wrote: > > Hi Robert, > > > > thanks for your answer. > > the aggregate function I was talking about is the function I need to use > for the non-group by columns like min() in my example. > > There are of course several function to