[TYPO3-performance] TYPO3 4.3 performance

[Rupert Germann]

hey certified intregator,

(OT: I'm interested in this, ähmmm,title, where can one get this
certification?)

Mathias Schreiber [wmdb >] wrote:
> Ingo Renner schrieb:
>> not necessarily as MySQL also has a cache. Especially with tests like ab
>> which is not comparable to real people behavior this could lead to false
>> results.
> 
> True.
> 
> It's like hitting the = button on a calculator over and over again ;-)
>
> a very simple method is 20-40 httracks all grabbing the full website
> (preferably via LAN in the datacenter) for half a day or so.

but what can httrack mesure?

I made some possibly more realistic tests than hitting the same button again
and again. 
"siege" has a handy option for this:
-i, --internet
  INTERNET, generates user simulation by randomly hitting the URLs read from
the urls.txt file.

see results here:
http://rgdata.de/fileadmin/user_upload/4.2.7_vs_4.3.0.pdf

greets
Rupert