Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 6 Nov 1996 16:21:28 -0700 (MST)
From:      Terry Lambert <terry@lambert.org>
To:        hasty@rah.star-gate.com (Amancio Hasty)
Cc:        archie@whistle.com, jgreco@brasil.moneng.mei.com, hannibal@cyberstation.net, hackers@freebsd.org
Subject:   Re: Limiting bandwidth on a socket? (SO_RCVBUF?)
Message-ID:  <199611062321.QAA09039@phaeton.artisoft.com>
In-Reply-To: <199611062125.NAA11085@rah.star-gate.com> from "Amancio Hasty" at Nov 6, 96 01:25:00 pm

next in thread | previous in thread | raw e-mail | index | archive | help
> What about kernel modifications?
> In a commercial environment it can  be very beneficial to provide rate
> control. QOS --- that ugly old OSI term or more recently reservation
> bandwith services.

That's only useful if your reservation is end-to-end, isn't it?

I mean if you have a T1, you know you have a T1, and you reserve 50%
of the thing, then you have to be sure that you will get that rate
end-to-end for it to be truly useful.


What you guys are really complaining about is "Bob is eating all my
bandwidth with his app, how do I throttle Bob?".

I think the answer is hidden in the question "how do I *know* Bob is
eating all my bandwidth in the first place?  Specifically, how do I
know how much I have so that 'all my bandwidth' is a meaningful idea?".


It's only useful to talk about 50% of X when you know how big X is...


					Regards,
					Terry Lambert
					terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199611062321.QAA09039>