Re: Tor client performance

[Olaf Selke <olaf.selke@xxxxxxxxxxxx>]

Andrew Lewman schrieb:
> On Monday May 10 2010 10:21:24 Olaf Selke wrote:
>> According torperf data from
>> http://metrics.torproject.org/csv/torperf.csv latency started to drop
>> 5th May:
>>
>> source		date		q1	md	q3
>> torperf-50kb	02.05.2010	7735	13092	20525
>> torperf-50kb	03.05.2010	9254	14550	20619
>> torperf-50kb	04.05.2010	7810	12498	20961
>> torperf-50kb	05.05.2010	6459	9848	15739
>> torperf-50kb	06.05.2010	4852	8069	14107
>> torperf-50kb	07.05.2010	4012	6036	10384
>> torperf-50kb	08.05.2010	3859	6080	9807
>> torperf-50kb	09.05.2010	3733	5766	9492
>> torperf-50kb	10.05.2010	3417	5073	6876
>>
>> Was it me?
> 
> Unofficially, I think no.  I would be concerned in one relay, as fast as 
> blutmagie's are, could have this much impact on the network.

in order to verify this today at 11:50 GMT+2 I restored the old state
before hardware upgrade by setting
BandBandwidthRate 14000 KB on blutmagie[1,2] and
BandBandwidthRate 100 KB on blutmagie[3,4].

This should overload two cores running blutmagie[1,2] processes leaving
two cores idle and put blutmagie[3,4] almost out of service.

Let's wait and look at http://metrics.torproject.org/torperf-graphs.html

Olaf