Re: 9.2beta1, parallel queries, ReleasePredicateLocks, CheckForSerializableConflictIn in the oprofile

On Sun, May 27, 2012 at 11:45 AM, Sergey Koposov <koposov(at)ast(dot)cam(dot)ac(dot)uk> wrote:
> Hi,
>
> I did another test using the same data and the same code, which I've
> provided before and the performance of the single thread seems to be
> degrading quadratically with the number of threads.
>
> Here are the results:
> Nthreads Time_to_execute_one_thread
> 1 8.1
> 2 7.8
> 3 8.1
> 4 9.0
> 5 10.2
> 6 11.4
> 7 13.3
> 8 16.1
> 9 19.0
> 10 21.4
> 11 23.8
> 12 27.3
> 13 30.2
> 14 32.0
> 15 34.1
> 16 37.5

This looks more skew hyperbolic than quadratic. In the higher region,
it doesn't seem to be increasing at an increasing rate, but rather
increasing at a constant rate of about 2.5 sec per additional backend,
i.e. converging to an asymptote with a slope of 2.5. A pedantic
point, perhaps, but if you are trying to infer the nature of the
bottleneck from the shape of the curve, they lead to rather different
sets of hypotheses.

Cheers,

Jeff