On Wed, Mar 21, 2007 at 01:40:52PM +0100, Daniel Lezcano wrote: > Dmitry Mishin wrote: > >On Wednesday 21 March 2007 12:47, Daniel Lezcano wrote: > > > >>Herbert Poetzl wrote: > >> > >>>On Tue, Mar 20, 2007 at 09:53:01PM +0100, Cedric Le Goater wrote: > >>> > >>>>All, > >>>> > >>>>We've been gathering, porting and testing a whole bunch of patchsets > >>>>related to namespaces, containers and resource management in what > >>>>we call the -lxc patchset. > >>>> > >>>great! > >>> > >>[ cut ] > >> > >> > >>>>* generic Process containers from Paul Menage <menage@xxxxxxxxxx> > >>>>* namespace entering from Serge E. Hallyn <serue@xxxxxxxxxx> > >>>>* resource controllers based on process containers from Pavel Emelianov > >>>><xemul@xxxxx> * multiple /proc (required for pid namespace) from Dave > >>>>Hansen <hansendc@xxxxxxxxxx> * pid namespace from Sukadev Bhattiprolu > >>>><sukadev@xxxxxxxxxx> > >>>>* L2 network namespace from Eric W. Biederman <ebiederm@xxxxxxxxxxxx> > >>>>* misc fixes and cleanups from others (sorry for not mentioning) > >>>> > >>>>and it's giving some good results on common platforms like i386 and > >>>>x86_64. > >>>> > >>>what _are_ the good results? do you have performance > >>>results or other interesting data on it? if so, where > >>>can it be found? > >>> > >>Hi Herbert, > >> > >>I played with the L2 namespace patchset from Eric Biederman, I did some > >>benchmarking with netperf: > >> > >>With 2 hosts, Intel EM64T bipro HT / 2,4 GHz , 4Go ram and GB network. > >>Host A is running the netserver on a RH4 kernel 2.6.9-42 > >>Host B is running the netperf client inside and outside the container > >>with the command: > >> netperf -H HostA -c -l 20 -n 2 -p 12865 > >> > >>Results are: > >>inside the container: > >> Throughput : 940.39 Mbit/s CPU usage : 15.80 % > >> > >>outside the container: > >> Throughput : 941.34 Mbits/s CPU usage : 5.80 % > >> > >Daniel, > > > >You probably did the same tests for my patchset also, didn't you? > >Which results did you get? > > > Effectively, I did some tests with your patchset but in a different way. > I did it with a bridge and with tbench so I didn't got the cpu usage and > througput is impacted. It will be irrelevant to give these values if we > can not compare them with Eric's patchset. > Anyway, you are right, it is interesting to have a comparison. For this > reason I added the ioctl in veth to facilitate automated benchmarking > and I am finishing the performances test suite for your patchset. I > will send the results ASAP. excellent, please don't forget to test with L3 isoltaion (a simple network namespace from Linux-VServer should do the trick) too, let me know if you have any issues with getting started ... TIA, Herbert _______________________________________________ Containers mailing list Containers@xxxxxxxxxxxxxxxxxxxxxxxxxx https://lists.linux-foundation.org/mailman/listinfo/containers