--sIsKpy5+/8vj5Foo
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Mon, Mar 22, 2004 at 10:00:18AM -0500, Thor Lancelot Simon wrote:
> Having looked at (and instrumented) the I/O spreading rotor as it is in
> your patch, I am not so sure it really does anything unless you happen
> to have disks that can clear well upwards of 8,192 metadata operations
> in a single second. =20

These numbers were determined by experiment, but the experiment wasn't
complete. As you know, I started with a rotor of (iirc) 32 ops per
second.  This caused VERY noticable performance degradation, and my
assumption was that this was because the drive (which has write-cache)
was no longer able to consolidate writes and was forced to seek much
more.

I then jumped right up to a much larger number, and came down until it
seemed to help. I haven't really explored the space between 8192 and
32 - but regardless of the value, magic numbers seem bad.

--
Dan.
--sIsKpy5+/8vj5Foo
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (NetBSD)

iD8DBQFAYhDKEAVxvV4N66cRAn3YAJ9pi7AipAPvaOCZFGu3gb6/mbGWnwCgyD6+
58aAMRuI7Ocs8qJWt6TORwI=
=8rE/
-----END PGP SIGNATURE-----

--sIsKpy5+/8vj5Foo--