Thread: Context Switching issue: Spinlock doesn't fix.

Folks,

I've been testing varying SPINS_PER_DELAY in a client's installation of
PostgreSQL against a copy of a production database, to test varying this
statistic as a way of fixing the issue.

It does not seem to work.

I've tested all of the following graduated levels:

100 (the original)
250
500
1000
2000
5000
10000
20000
30000
50000

None of these quantities seem to make any difference at all in the number of
context switches -- neither down nor up.   Seems to me like this is a dead
end.   Does anyone have test results that show otherwise?

--
-Josh Berkus
 Aglio Database Solutions
 San Francisco

Hello,

Is context switching problem resolved in 8.0?

Can I drop in another Xeon?

Thanks,
Jelle


On Wed, 2 Jun 2004, Josh Berkus wrote:

> Folks,
>
> I've been testing varying SPINS_PER_DELAY in a client's installation of
> PostgreSQL against a copy of a production database, to test varying this
> statistic as a way of fixing the issue.
>
> It does not seem to work.
>
> I've tested all of the following graduated levels:
>
> 100 (the original)
> 250
> 500
> 1000
> 2000
> 5000
> 10000
> 20000
> 30000
> 50000
>
> None of these quantities seem to make any difference at all in the number of
> context switches -- neither down nor up.   Seems to me like this is a dead
> end.   Does anyone have test results that show otherwise?
>
>

--

http://www.jibjab.com

Jellej,

> Is context switching problem resolved in 8.0?
>
> Can I drop in another Xeon?

Nope, not solved yet.   However, it only affects certain data access patterns.
So don't use it as a reason not to go multi-processor.

--
--Josh

Josh Berkus
Aglio Database Solutions
San Francisco