Re: [Ocfs2-devel] Ocfs2 performance

[Daniel Phillips <phillips@google.com>]

Going back to that mysterious idle time fluctuation problem, I just noticed
something interesting here:

> With hashvec, (64K buckets)
>
>       real user sys
>     29.62 23.87 3.02
>     28.80 24.16 3.07
>     50.95 24.11 3.04
>     28.17 23.95 3.20
>     61.21 23.89 3.16
>     28.61 23.88 3.30
>
> With vmalloc (64K buckets)
>
>       real user sys
>     29.67 23.98 2.88
>     28.35 23.96 3.13
>     52.29 24.16 2.89
>     28.39 24.07 2.97
>     65.13 24.08 3.01
>     28.08 24.02 3.16

Look, the real time artifacts show up in the same places in the two
different runs.  There was a reboot before each run.  There is a sync
before  and after each trial.  Each trial was initiated by hand, so this
isn't about something else going on in the machine at the same time.

Got to be a clue in there somewhere.

Regards,

Daniel
-
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/