On Fri, 21 Dec 2001, Mike Kravetz wrote:
> On Fri, Dec 21, 2001 at 09:19:04AM -0800, Davide Libenzi wrote:
> > On Fri, 21 Dec 2001, Mike Kravetz wrote:
> >
> > > Some time back, I asked if anyone had any RT benchmarks and got
> > > little response. Performance (latency) degradation for RT tasks
> > > while implementing new schedulers was my concern. Does anyone
> > > have ideas about how we should measure/benchmark this? My
> > > 'solution' at the time was to take a scheduler heavy benchmark
> > > like reflex, and simply make all the tasks RT. This wasn't very
> > > 'real world', but at least it did allow me to compare scheduler
> > > overhead in the RT paths of various scheduler implementations.
> >
> > Mike, a better real world test would be to have a variable system runqueue
> > load with the wakeup of an rt task and measuring the latency of the rt
> > task under various loads.
> > This can be easily implemented with cpuhog ( that load the runqueue ) plus
> > the LatSched ( scheduler latency sampler ) that will measure the exact
> > latency in CPU cycles.
>
> Right! Any ideas on variable system runqueue load? Should those
> other tasks be RT or OTHER? a mix? I would suspect that we would
> want multiple RT tasks on the runqueue or at least in the system
> (otherwise why worry about global semantics?).
>
> However, I would feel better about this if someone had a real world
> load involving RT tasks on a SMP system. At least then we could try
> to simulate a load someone cares about.
In my tests i stop the run queue load to 8 ( per cpu ) now coz higher
values are somehow unusual.
A good plot should also have a third dimension that is the number of real
time tasks running.
I guess i've to take a better look at gnuplot docs for 3d graphs :)
- Davide
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sun Dec 23 2001 - 21:00:25 EST