> 1) Running Oracle apps, doing raw IO. We are running an -aa kernel
> on this machine, and it doesn't help.
> So, we haven't really given Andrea's patch a fair test yet. If you
> guys can agree which the better approach is by just discussing it,
One thing that seems to be missing in Linux are proper VM
statistics. There is no way handwaving and discussions are
going to be better than a measurement of what's going on
inside the VM.
Treating the different VM patch sets as black boxes with
benchmarks will show us which VM works best for which
benchmark, but it won't show us why or how to combine the
good features of the different VMs...
I think good statistics to start would be the traditional
page fault rate (pf), page free rate (fr), page scan rate
(sr), reclaims (re), pageout (po), pagein (pi) and some
variation of iowait stats.
regards,
Rik
-- Bravely reimplemented by the knights who say "NIH".http://www.surriel.com/ http://distro.conectiva.com/
- 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/