[188903] in North American Network Operators' Group
Re: Latency, TCP ACKs and upload needs
daemon@ATHENA.MIT.EDU (Eygene Ryabinkin)
Fri Apr 22 02:18:01 2016
X-Original-To: nanog@nanog.org
Date: Fri, 22 Apr 2016 09:17:54 +0300
From: Eygene Ryabinkin <rea+nanog@grid.kiae.ru>
To: nanog@nanog.org
In-Reply-To: <20160420142753.GA58668@ussenterprise.ufp.org>
Errors-To: nanog-bounces@nanog.org
--TdkiTnkLhLQllcMS
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
Wed, Apr 20, 2016 at 07:27:53AM -0700, Leo Bicknell wrote:
> 90%+ of the stacks deployed will be too small. Modern Unix generally
> has "autotuning" TCP stacks, but I don't think Windows or OS X has
> those features yet
OS X since ~10.5 has autotuning, here are some hints from ESnet
https://fasterdata.es.net/host-tuning/osx/
Personally never tried this on the sat-type RTT. As explained,
scaling factor of 3 is a limiting one for high-performance transfers.
For sat links may limit the things too, since 64K * 2^3 / 0.9 ms RTT * 8 ~
4.5 Mbit/sec, but one's mileage may vary. And, of course, packet loss
can turn down this BDP-derived speed drastically.
--=20
Eygene Ryabinkin, National Research Centre "Kurchatov Institute"
Always code as if the guy who ends up maintaining your code will be
a violent psychopath who knows where you live.
--TdkiTnkLhLQllcMS
Content-Type: application/pgp-signature; name="signature.asc"
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2
iKYEABECAGYFAlcZwg9fFIAAAAAALgAoaXNzdWVyLWZwckBub3RhdGlvbnMub3Bl
bnBncC5maWZ0aGhvcnNlbWFuLm5ldDZDNjExQkNFMUNCMTI1NDhGQTdGQzI0REFD
Q0UwMTJDNUM4QjdFQzcACgkQrM4BLFyLfsc1KACeKAEm5D5pCm9P4HuOGrodxN9H
6ekAniFCX72C7VcA6ot2GjLLNIzvqaz2
=1p3r
-----END PGP SIGNATURE-----
--TdkiTnkLhLQllcMS--
