IETF Logo Mail Archive

Re: types of traffic in tcp congest control

Joe Touch <touch@ISI.EDU> Mon, 13 May 2002 18:46 UTC

Message-ID: <3CE009FA.7080200@isi.edu>
Date: Mon, 13 May 2002 11:46:18 -0700
From: Joe Touch <touch@ISI.EDU>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.0rc1) Gecko/20020417
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: J Wu <jinw@comp.leeds.ac.uk>
Cc: =?ISO-8859-1?Q?=B4=F3=C0=EE?= <liy666@sina.com>, tcp-impl@lerc.nasa.gov
Subject: Re: types of traffic in tcp congest control
References: <Pine.LNX.4.33.0205131918000.30118-100000@cslin135.leeds.ac.uk>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Sender: owner-tcp-impl@grc.nasa.gov
Precedence: bulk
Status: RO
Content-Length: 699
Lines: 25

J Wu wrote:
> On Mon, 13 May 2002, Joe Touch wrote:
> 
> 
>>Only if there _are_ elephants. In the transatlantic case, a huge number
>>of mice can swamp the connection - yes, a severe shortage. One solution
>>is to "RED" the SYN packets of new connections, or limit how many
>>connections can be established.
> 
> In this case, UDP "mice" will still congest the network.

UDP isn't a mice or elephant, since it won't backoff anyway.

 > Also, at the
> angel of core switch, how can it decide mice from elephant?

Generally it can't.

> And, is it technically possible to trace the TCP numbers at core switch?

Possible, yes. Feasible, no - the number of connections gets too large 
too fast...

Joe

v2.8.7 | Report a Bug | By Email