[Devel] Re: Pid namespaces approaches testing results

[Pavel Emelianov]

Dave Hansen wrote:
> On Tue, 2007-05-29 at 15:45 +0400, Pavel Emelianov wrote:
>> The detailed results are the following:
>> Test name:    spawn     execl    shell    ps (sys time)
>> 1(no ns) :    579.1     618.3    1623.2   3.052s
>> 2(suka's):    570.7     610.8    1600.2   3.107s
>> Slowdown :    1.5%      1.3%     1.4%     1.8%
>>
>> 3(no ns) :    580.6     616.0    1633.8   3.050s
>> 4(flat)  :    580.8     615.1    1632.2   3.054s
>> Slowdown :    0%        0.1%     <0.1%    0.1%
>> 5(multi) :    576.9     611.0    1618.8   3.065s
>> Slowdown :    0.6%      0.8%     0.9%     0.5%
> 
> Wow, thanks so much for running those.  You're a step ahead of us,
> there!

Thanks :) Maybe we shall cooperate then and make three series
of patches like

1. * The Kconfig options;

   * The API. I.e. calls like task_pid_nr(), task_session_nr_ns() etc;
   This part is rather important as I found that some places in kernel
   where I had to lookup the hash in multilevel model were just pid->vpid
   dereference in flat model. This is a good optimization.

   * The changes in the generic code that intruduce a bunch of 
   #ifdef CONFIG_PID_NS
    ...
   #else
   #ifdef CONFIG_PID_NS_FLAT
   #endif
   #ifdef CONFIG_PID_NS_MULTILEVEL
   #endif
   #endif
   code in pid.c, sched.c, fork.c etc

   This patchset will have to make kernel prepared for namespaces injections
   and (!) not to break normal kernel operation with CONFIG_PID_NS=n.

2. The flat pid namespaces (my part)
3. The multilevel pid namespaces (suka's part)

> Did you happen to collect any profiling information during your runs? 

Unfortunately no :( My intention was to prove that hierarchy has
performance implications and should be considered carefully.

> -- Dave
> 
> 

_______________________________________________
Containers mailing list
Containers at lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers