Re: AIX slow buffer reads

----- 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 times.
| >
|
| Brad's point is that it probably doesn't do anything at all on AIX,
| and is already disabled accordingly. But on Linux, it is doing something,
| and that might be contributing to why it's executing so much better on
| that platform. If you disable that parameter on your Debian box, that
| should give you an idea whether that particular speed-up is a major
| component to the difference you're seeing or not.

Cant do it right now, but will do it ASAP and post here.

| Also, if the system check was done by the "vendor team" team, don't
| trust them at all. It doesn't sound like a disk problem is involved in
| your case yet, but be sure to do your own basic disk benchmarking too
| rather than believing what you're sold. There's a quick intro to that
| at
| http://www.westnet.com/~gsmith/content/postgresql/pg-disktesting.htm
| and a much longer treatment of the subject in my book if you want a
| lot
| more details. I don't have any AIX-specific tuning advice in there
| though.|

I´m gonna read your sugestion, thanks.
We tested the disks also, and we did a lot of tuning to get acceptable transfer rates at AIX.

Yesterday I tried your "stream-scaling" and get around 7000MB/s (single thread) and 10000MB/s (eight threads) at AIX, and a little less than that at Debian, since its a virtual box.
I found that even my notebook is close to that transfer rates, and both boxes are limited by DDR2 speeds.

| --
| Greg Smith 2ndQuadrant US greg(at)2ndQuadrant(dot)com Baltimore, MD
| PostgreSQL Training, Services and Support www.2ndQuadrant.us
| "PostgreSQL 9.0 High Performance": http://www.2ndQuadrant.com/books

[]´s, André Volpato