Re: [patch v3]numa: add a sysctl to control interleave allocationgranularity from each node to improve I/O performance

[Shaohua Li]

On Thu, 2011-12-15 at 01:53 +0800, Andi Kleen wrote:
> > That's what I want to avoid letting each apps to explicitly do it, it's
> > a lot of burden.
> 
> Usually apps that set NUMA policy can change it. Most don't anyways.
> If it's just a script with numactl it's easily changed.
Hmm, why should apps set different granularity? the granularity change
is to speed up I/O, which should have the same value for all apps.

> > That's true only workload with heavy I/O wants this. but I don't expect
> > it will harm other workloads.
> 
> How do you know?
I can't imagine how it could harm. Some arches can use big pages, big
granularity should already been tested for years.

Thanks,
Shaohua

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