QoS per TCP connection

Harald Welte laforge@gnumonks.org
Mon, 1 Oct 2001 16:39:36 +0200


On Sun, Sep 30, 2001 at 02:17:55PM +0300, Sampsa Ranta wrote:
> 
> Hi!
> 
> You wrote that piece of code to track TCP window.. Do you by any
> change have a piece of code that allows adjustments to that window size.
> 
> I just learned (in university, something good comes out of those, too =)
> that window size is the thing that tells us how much data there can be in
> transit and adjusting it could be used as to limit bandwidth used by a
> single TCP connection atleast on link with known bandwidth.
> 
> In other words, this could be used as very simple shaper for a single TCP
> connection, which would be good and simple way.. In case we also had
> byte counter in conntrack entry, it would be possible to start limiting
> data rate if the TCP connection was used to transfer data enough.
> 
> What do you think?

yes, this is a well-known mechanism for throttling tcp connections, and it
is deployed in several commercially available products.

It is, however, to be doubted that this severe mangling of TCP header data
is really desirable on an intermediate router.  Every step we do in this
direction further hurts end-to-end transparancy.

We already have NAT and other evil things - lets try not to invent more of 
them...

>  - Sampsa Ranta
>    sampsa@netsonic.fi

-- 
Live long and prosper
- Harald Welte / laforge@gnumonks.org               http://www.gnumonks.org/
============================================================================
GCS/E/IT d- s-: a-- C+++ UL++++$ P+++ L++++$ E--- W- N++ o? K- w--- O- M- 
V-- PS+ PE-- Y+ PGP++ t++ 5-- !X !R tv-- b+++ DI? !D G+ e* h+ r% y+(*)