Re: stress testing

[Paul Syverson <syverson@itd.nrl.navy.mil>]

I talked to Roger about this a bit. He pointed out that it would be
nice to have a facility that he, Mat, etc. could play with directly. On
the other hand this sounds like a useful resource likely to yield
useful results.
Bottom line: go ahead and set things up.

-Paul

> Date: Thu, 11 Jul 2002 11:12:53 -0400
> From: Bruce Montrose <montrose@itd.nrl.navy.mil>
> 
> We have stress-test software called WebLoad and many machines all setup
> to do this kind of testing in our lab. The software and machines belong to
> a different section so I will need to play some politics to arrange for our 
use
> of these resources for the onion router.
> 
> Please give me some feedback regarding this, since I do not want to do this
> if there is no interest on our end to use these resources for our 
> stress-testing.
> 
> -Bruce
> 
> At 08:42 AM 7/11/2002 -0400, you wrote:
> >We have some folks over here at NRL that have software to stress-test
> >network connections. I will talk to them later today since they are not here
> >yet to find out if it could be used to test the onion router.
> >
> >At 03:59 AM 7/11/2002 -0400, you wrote:
> >
> >>I am going to start writing a "network simulation", i.e. something that
> >>will make many connections through the or net and verify that the transfer
> >>was OK. Whatever bugs there might still be in there, I reckon they'll show
> >>up if we use something like that ...
> >>
> >>First of all, has anyone already written anything like that and has hidden
> >>it from the allseeing cvs repository? ;-)
> >>
> >>I am also accepting feature requests - for now I'll just verify the data
> >>transfer. Delay would also be a good thing to measure, anything else?
> >>
> >>Thanks,
> >>Mat
> >>
> >>--
> >>Matej Pfajfar
> >>
> >>GPG Public Keys @ http://matejpfajfar.co.uk/keys