Linux scheduler, overscheduling performance, threads

[Brian Hurt]

I've given this thread up.  While I'm not convinced that there isn't a
problem, I am convinced that most people on this list are convinced that
there isn't.  When/if it bites me, I'll write a patch and have real code
to show people.  If it bites IBM before it bites me and they fix it, well
and good.  So in a sense it isn't a problem- if it becomes one it can (and
will) get fixed.

On Mon, 24 Jan 2000, Ingo Molnar wrote:

> 
> On Thu, 20 Jan 2000, Brian Hurt wrote:
> 
> > Following this thread off and on since it's inception, and being a Java
> > programmer myself, can I offer some observations?
> > 
> > Thousands of threads in a program is not unreasonable.  If you may want to
> > take full advantage of a 128 CPU machine, for example, you need _at_
> > _least_ 128 threads.  If your threads spend most of their time blocking,
> > you need even more threads, you need to overschedule, to make sure you
> > generally have enough threads not blocking to make sure CPUs aren't going
> > to waste.  Unfortunately, due to vagaries of the system, you will have
> > points when most of the threads become runnable at once.
> 
> i believe you are confusing '2000 threads created and happily waiting
> blocked to do something' and the '2000 threads running' case. I
> specifically included a quick benchmark showing 20 thousand threads being
> around on my system and not impacting Linux scheduler performance the
> slightest. What i say is that more than nr_cpus*2 _running_ (ie. currently
> waiting for the CPU, not some other resource) threads are bad.
> 
> -- mingo
> 
> 


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo en vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/