[Benchmarking] Last minute vote: shall we switch to 6600requestsrun?

[Adrian Custer]

Hello,

On Thu, 2010-09-02 at 18:45 +0200, Anne-Sophie Collignon wrote:
> 
> What do you think about the following alternative ?
> 
> The idea is to publish the results of the 3 runs in the final
> presentation, and to provide comments on the possible differences that
> could be observed on the 3 runs results.
> 

I am confused by your proposal. You say "publish the results of the 3
runs". 

That might mean publish all three sets of runs from the newer scripts or
might mean publish only the last runs (1,2,4,8,16,32,64 threads) from
the newer scripts, I am not sure. From what I understand, Andrea's
strategy is to report only the last set of any run, with the earlier
runs intended merely to 'warm up' the server in equivalent ways, either
with the same requests or with different requests.


Also, are you saying 
  The idea is to publish {only} the ...
or 
  The idea is to publish {also} the ...
?

I am wondering mostly if I should run the first set of tests as well as
the newer set of runs.

thanks,
  --adrian