Re: [PERFORM] Article about PostgreSQL and RAID in Brazil

2004-09-16 Thread Daniel Ceregatti
The first part of the article can be found in http://www.sqlmagazine.com.br/colunistas.asp?artigo=Colunistas/RicardoRezende/06_Raid_P1.asp The site seems to be down. I was looking forward to reading it. :( The first and most important step for RAID performance with PostgreSQL is to get a

Re: [PERFORM] Article about PostgreSQL and RAID in Brazil

2004-09-16 Thread Steinar H. Gunderson
On Thu, Sep 16, 2004 at 11:10:13AM -0700, Daniel Ceregatti wrote: Here is a small example of the performance difference with write cache: http://sh.nu/bonnie.txt Am I missing something here? I can't find any tests with the same machine showing the difference between writeback and

Re: [PERFORM] Article about PostgreSQL and RAID in Brazil

2004-09-16 Thread Josh Berkus
Primer, The site seems to be down. I was looking forward to reading it. :( I didn't have a problem. The site *is* in Portuguese, though. -- --Josh Josh Berkus Aglio Database Solutions San Francisco ---(end of broadcast)--- TIP 8: explain

Re: [PERFORM] Article about PostgreSQL and RAID in Brazil

2004-09-16 Thread Daniel Ceregatti
Josh Berkus wrote: Primer, The site seems to be down. I was looking forward to reading it. :( I didn't have a problem. The site *is* in Portuguese, though. Yes, it came up finally. Fortunately I'm Brazilian. :) -- Daniel Ceregatti - Programmer Omnis Network,

Re: [PERFORM] Article about PostgreSQL and RAID in Brazil

2004-09-16 Thread Qing Zhao
Hi, there, I am running PostgreSQL 7.3.4 on MAC OS X G5 with dual processors and 8GB memory. The shared buffer was set as 512MB. The database has been running great until about 10 days ago when our developers decided to add some indexes to some tables to speed up certain uploading ops. Now the CPU

Re: [PERFORM] Article about PostgreSQL and RAID in Brazil

2004-09-16 Thread Jim C. Nasby
On Thu, Sep 16, 2004 at 10:50:33AM -0700, Josh Berkus wrote: The second step is to have lots of disks; 5 drives is a minimum for really good performance. 3-drive RAID5, in particular, is a poor performer for PostgreSQL, often resulting in I/O that is 40% or less as efficient as a single

Re: [PERFORM] Article about PostgreSQL and RAID in Brazil

2004-09-16 Thread Josh Berkus
Jim, What about benefits from putting WAL and pg_temp on seperate drives? Specifically, we have a box with 8 drives, 2 in a mirror with the OS and WAL and pg_temp; the rest in a raid10 with the database on it. Do you think it would have been better to make one big raid10? What if it was

Re: [PERFORM] Article about PostgreSQL and RAID in Brazil

2004-09-16 Thread Jim C. Nasby
On Thu, Sep 16, 2004 at 02:07:37PM -0700, Josh Berkus wrote: Jim, What about benefits from putting WAL and pg_temp on seperate drives? Specifically, we have a box with 8 drives, 2 in a mirror with the OS and WAL and pg_temp; the rest in a raid10 with the database on it. Do you think it