Re: [PERFORM] AIX slow buffer reads

2010-10-28 Thread André Volpato
- Mensagem original - | André Volpato wrote: | | | | If it is being spent in the bitmap index scan, try setting | | effective_io_concurrency to 0 for Linux, and see what effect that | | has. | | I disabled effective_io_concurrency at AIX but it made no changes on | bitmap index

Re: [PERFORM] AIX slow buffer reads

2010-10-27 Thread Brad Nicholson
On 10-10-26 05:04 PM, André Volpato wrote: - Mensagem original - | On 10-10-25 03:26 PM, André Volpato wrote: | | On Mon, Oct 25, 2010 at 2:21 PM, André Volpato | |andre.volp...@ecomtecnologia.com.br wrote: (...) | | These times keep repeating after the second run, and I can | |

Re: [PERFORM] AIX slow buffer reads

2010-10-27 Thread Greg Smith
André Volpato wrote: | | If it is being spent in the bitmap index scan, try setting | effective_io_concurrency to 0 for Linux, and see what effect that has. I disabled effective_io_concurrency at AIX but it made no changes on bitmap index times. Brad's point is that it probably doesn't

Re: [PERFORM] AIX slow buffer reads

2010-10-27 Thread Tom Lane
Greg Smith g...@2ndquadrant.com writes: André Volpato wrote: I disabled effective_io_concurrency at AIX but it made no changes on bitmap index times. Brad's point is that it probably doesn't do anything at all on AIX, and is already disabled accordingly. AFAICT from googling, AIX does

Re: [PERFORM] AIX slow buffer reads

2010-10-27 Thread André Volpato
- Mensagem original - | André Volpato wrote: | | | | If it is being spent in the bitmap index scan, try setting | | effective_io_concurrency to 0 for Linux, and see what effect that | | has. | | I disabled effective_io_concurrency at AIX but it made no changes on | bitmap index

Re: [PERFORM] AIX slow buffer reads

2010-10-27 Thread Brad Nicholson
On 10/27/2010 4:10 PM, Tom Lane wrote: Greg Smithg...@2ndquadrant.com writes: André Volpato wrote: I disabled effective_io_concurrency at AIX but it made no changes on bitmap index times. Brad's point is that it probably doesn't do anything at all on AIX, and is already disabled

Re: [PERFORM] AIX slow buffer reads

2010-10-27 Thread Tom Lane
Brad Nicholson bnich...@ca.afilias.info writes: On 10/27/2010 4:10 PM, Tom Lane wrote: AFAICT from googling, AIX does have posix_fadvise, though maybe it doesn't do anything useful ... If there is an easy way to check if it does do anything useful? If so, I can check it out. If you don't

Re: [PERFORM] AIX slow buffer reads

2010-10-26 Thread Brad Nicholson
On 10-10-25 03:26 PM, André Volpato wrote: | On Mon, Oct 25, 2010 at 2:21 PM, André Volpato |andre.volp...@ecomtecnologia.com.br wrote: | Hi all, | | We are tuning a PostgreSQL box with AIX 5.3 and got stucked in a | very odd situation. | When a query got ran for the second time, the system

Re: [PERFORM] AIX slow buffer reads

2010-10-26 Thread André Volpato
- Mensagem original - | On 10-10-25 03:26 PM, André Volpato wrote: | | On Mon, Oct 25, 2010 at 2:21 PM, André Volpato | |andre.volp...@ecomtecnologia.com.br wrote: (...) | | These times keep repeating after the second run, and I can | | ensure AIX isn´t touching the disks anymore.

Re: [PERFORM] AIX slow buffer reads

2010-10-25 Thread André Volpato
| On Mon, Oct 25, 2010 at 2:21 PM, André Volpato | andre.volp...@ecomtecnologia.com.br wrote: | Hi all, | | We are tuning a PostgreSQL box with AIX 5.3 and got stucked in a | very odd situation. | When a query got ran for the second time, the system seems to | deliver the results to slow. |