Re: Fwd: dm perf scales poorly with 2.6.x kernel

[Andi Kleen]

Mike Snitzer <snitzer@xxxxxxxxx> writes:
>
> Ultimately there appears to be a disproportionate increase in the DM
> performance hit as the IO throughput capability of the underlying
> block device increases (I'm obviously using a limited sampling so...).
>  But given the current results it is clear there will be a performance
> hit associated with using DM; but having that hit be fixed as the
> throughput scales would be ideal.

You could do a oprofile run to see where to CPU time is going to
and post the results.

-Andi
-
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/