Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 20 Jan 2000 11:13:04 +1300
From:      Joe Abley <jabley@patho.gen.nz>
To:        Damian Hasak <dhasak@fore.com>
Cc:        David Wolfskill <dhw@whistle.com>, freebsd-isp@FreeBSD.ORG, xiyuan@yahoo.com
Subject:   Re: net speed
Message-ID:  <20000120111303.B22700@patho.gen.nz>
In-Reply-To: <003c01bf62a5$ec375b80$148190a9@dhasak-pc.fore.com>; from dhasak@fore.com on Wed, Jan 19, 2000 at 12:52:16PM -0500
References:  <20000119195314.A4167@patho.gen.nz> <003c01bf62a5$ec375b80$148190a9@dhasak-pc.fore.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, Jan 19, 2000 at 12:52:16PM -0500, Damian Hasak wrote:
> hmm but i would think jitter is more related to timing on a CBR connection

Can't a noticable variation in delay in successive segments (with
the timestamp option, for example) result in a wildly varying transmit
window size on both sides?

Somewhat implementation-specific, certainly, but in general TCP
behaves much more efficiently if the round-trip time is reasonable
predictable (or poorly measured, alternatively, as long as the poor
measurement is reasonable :)

Virtual-circuit concerns in ATM mirror session concerns in TCP to some
degree, except that congestion conditions are identified (and remedied)
in somewhat different ways.


Joe


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-isp" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20000120111303.B22700>