Re: 2 cents re: MTU

Jason Williams (streak@ccwf.cc.utexas.edu)
Thu, 25 Jun 1998 10:37:13 -0500 (CDT)


On Wed, 24 Jun 1998, Jamie Erbes wrote:
> MTU - Max transmission unit - default from windows is 1500 which is good
> for LAN but typically not so good for analog dial-up. The objective is
> to match the setting of your ISP, which is usually 576.

So the question becomes...why would an ISP force you to use an MTU of 576?
And would it really matter if your ISP uses an MTU of >1500? The way I've
always thought is that most routers support an MTU of 1500. You want the
largest packet size allowed without fragmentation. If ONLY the end points
are using an MTU of 576, it seems like a lot is wasted. From what I know,
fewer bigger packets is always more desirable than a lot of tiny packets
as long as the bigger packets aren't fragmented into smaller pieces.

> RWIN - Receive window - I've seen Windows default to 8192, which, again,
> is OK for LAN but not so good for analog dial-up. A good rule of thumb
> is to set RWIN at 4X your MSS. Then round it to a number divisible by
> 32 for further optimization.

Perhaps I need to read up some more, but I believe RWIN and MSS are TCP
specific. CU-SeeMe uses UDP which doesn't have any "Receive window"
(which I believe is the reason the reflector can easily flood a modem off
and backlog it with packets). TCP's sliding window protocol is designed
to handle that...UDP isn't. RWIN is defined as the number of bytes that
can be received at the beginning of the connection before it can be
increased or decreased? Is that the sliding window? How does that effect
UDP? The IP layer has no control over congestion and neither does UDP as
far as I know.

--
streak@ccwf.cc.utexas.edu    * Jason Williams -- Austin, Tx.  |     |
streak@mail.utexas.edu       * University of Texas at Austin  | ___ |
streak@cs.utexas.edu         * BS Computer Science             \_|_/
*************** http://ccwf.cc.utexas.edu/~streak/ **************|