Re: [Lse-tech] Re: [patch] scheduler fix for 1cpu/node case

[Andrew Theurer <habanero@us.ibm.com>]

> >AMD is 1 because there's no need to balance within a node, so I want the
> >inter-node balance frequency to be as often as it was with just O(1). 
> > This interval would not work well with other NUMA boxes, so that's the
> > main reason to have arch specific intervals.
>
> OK, I misread the patch. IIRC AMD has 1 CPU per node? If so, why doesn't
> this simply prevent balancing within a node?

Yes, one cpu/node.  Oh, it does prevent it, but with the current intervals, we 
end up not really balancing as often (since we need a inter-node balance), 
and when we call load_balance in schedule when idle, we don't balance at all 
since it's only a node local balance.

> >  And, as a general guideline, boxes with
> >different local-remote latency ratios will probably benefit from different
> >inter-node balance intervals.  I don't know what these ratios are, but I'd
> >like the kernel to have the ability to change for one arch and not affect
> >another.
>
> I fully appreciate there are huge differences... I am curious if
> you can see much improvements in practice.

I think AMD would be the first good test.  Maybe Andi has some results on 
numasched vs O(1), which would be a good indication.