[137817] in North American Network Operators' Group
Re: New transit tests
daemon@ATHENA.MIT.EDU (Mikael Abrahamsson)
Sat Feb 19 23:39:06 2011
Date: Sun, 20 Feb 2011 05:38:57 +0100 (CET)
From: Mikael Abrahamsson <swmike@swm.pp.se>
To: Diogo Montagner <diogo.montagner@gmail.com>
In-Reply-To: <AANLkTi=WP7fhP31_ncAPwJFaKJDZoXhx6k5evNzLdfg5@mail.gmail.com>
Cc: nanog@nanog.org
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
On Sun, 20 Feb 2011, Diogo Montagner wrote:
> Your test will be limited by the latency. You can do some tweak in the
> TCP but it will not improve too much considering the latency between BR
> and US.
With 200ms delay he only needs 1 megabyte of TCP window to reach 40
megabit/s. This is definitely not out of reach in case the transport is
loss free.
The world speed record a few years ago was approximately 10 gigabit/s in a
single TCP session, I don't know if someone has bested that yet.
--
Mikael Abrahamsson email: swmike@swm.pp.se
