Date: Mon, 8 Nov 2004 22:04:39 +0100 From: Emanuel Strobl <Emanuel.Strobl@gmx.net> To: freebsd-current@freebsd.org Cc: Robert Watson <rwatson@freebsd.org> Subject: Re: asymmetric NFS transfer rates Message-ID: <200411082204.41913.Emanuel.Strobl@gmx.net> In-Reply-To: <200411080429.12846.Emanuel.Strobl@gmx.net> References: <Pine.NEB.3.96L.1041102131322.21044C-100000@fledge.watson.org> <20041102105534.K63929@carver.gumbysoft.com> <200411080429.12846.Emanuel.Strobl@gmx.net>
next in thread | previous in thread | raw e-mail | index | archive | help
--nextPart8801253.ULrAexxBJs
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
Am Montag, 8. November 2004 04:29 schrieb Emanuel Strobl:
> Am Dienstag, 2. November 2004 19:56 schrieb Doug White:
> > On Tue, 2 Nov 2004, Robert Watson wrote:
> > > On Tue, 2 Nov 2004, Emanuel Strobl wrote:
[...]
> Now back to my real problem: Can you imagine that NFS and twe are blocking
> each other or something like that? Why do I get such really bad transfer
> rates when both parts are in use but every single part on its own seems to
> work fine?
I don't think so but by chance, can it have todo with this bug (from TWE(4)):
BUGS
The controller cannot handle I/O transfers that are not aligned to a
512-byte boundary. In order to support raw device access from user-
space, the driver will perform alignment fixup on non-aligned data. This
process is inefficient, and thus in order to obtain best performance
user-space applications accessing the device should do so with aligned
buffers.
Thanks,
-Harry
>
> Thanks for any help,
>
> -Harry
--nextPart8801253.ULrAexxBJs
Content-Type: application/pgp-signature
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.6 (FreeBSD)
iD8DBQBBj99pBylq0S4AzzwRAqbjAJ9651gXcQ/XFQW1WUhz0pqr9cnQ5QCgj4iq
QxC3wu3ALzxgOpW2qoOsaEc=
=7PyF
-----END PGP SIGNATURE-----
--nextPart8801253.ULrAexxBJs--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200411082204.41913.Emanuel.Strobl>