MTU of the Internet?

Several people have noted to the Microsoft Support and Product groups that
they want the Windows 95 PPP MTU to be set to 576 (down from 1500). this
change is in Windows 98.

The reason for this change cited by many customers is that many ISPs have
576 MTUs set "inside" their networks and packets get fragmented.

How prevelant is this fragmentation and how prevelant is this MTU in ISPs?

Are there statistics that can be shared on how much traffic they see on
their networks that are IP fragments?

Why do ISPs set their MTUs to 576 instead of ~1500 or even ~4K?

thanks in advance for your responses,

peter

The reason for this change cited by many customers is that many ISPs have
   576 MTUs set "inside" their networks and packets get fragmented.

   How prevelant is this fragmentation and how prevelant is this MTU in ISPs?

Unless your ISP uses BBN Butterflies and C30 IMPs in its backbone, I
would discount the odds of running into a link with an MTU of 576.

   Are there statistics that can be shared on how much traffic they see on
   their networks that are IP fragments?

   Why do ISPs set their MTUs to 576 instead of ~1500 or even ~4K?

Setting your MSS to higher than the MTU on any network over which the
packet will be routed will guarantee fragmentation. That's why people
generally don't set it higher than 1500, which is the MTU on ethernet,
Cisco serial lines, and a lot of other media (most of which can be
attributed to inheritence from Ethernet). FDDI and HSSI interfaces
are generally set to 4470 unless someone throttles it back.

I have no idea where they came up with this "576 internally" nonsense.
Generally whenever one runs into that number it is as a result of
creaky old software that expects to be running over milnet or arpanet.

Are Microsoft stacks known to be broken in the packet
fragmentation/reassembly department? Or are just acknowledging
deficiencies in their path mtu discovery code by setting the MSS in
the basement? I knew they had problems with window length (this from
my friends with long fat pipes)...

                                        ---Rob

Several people have noted to the Microsoft Support and Product groups that
they want the Windows 95 PPP MTU to be set to 576 (down from 1500). this
change is in Windows 98.

Yuck. Why bother implementing PTMU discovery if you are going to use a
MTU that larger than almost every MTU around?

The reason for this change cited by many customers is that many ISPs have
576 MTUs set "inside" their networks and packets get fragmented.

I really don't buy that. Many or most backbone links have MTU >1500, and
MTUs <1500 outside of low-speed dialup connections aren't that common.
They are there, yes. But not that common.

My understanding of why a lower MTU is demonstratable better under Win95
is because the Win95 TCP stack is broken, and it is a good workaround.
Most of the people raving about it are saying they are getting 2-4 times
speed increases from changing their MTU from 1500 to 576. Something is
very wrong there. I thought I had heard details about exactly what is
broken in the Win95 TCP stack that causes this problem, but can't recall
them at the moment. It could have no basis in reality and just be a
rumour.

There are all sorts of people spouting all sorts of lies around Windows
newsgroups about why small MTUs are good; I think novice users are simply
getting drawn in by supposed experts.

I guess systems receiving data from servers with broken retransmission
timers (eg. how Solaris used to be) could be helped by lowering the MTU
which would result in faster ACKs so bogus retransmissions won't happen
all the time, but the fix for this really isn't to lower the MTU.

You also get the obvious improvements in interactive performance, and you
start getting data more quickly.

I would suggest that you would be well advised to find a handy user or
four where this effect is easily observable, and find out what is really
going on.

> The reason for this change cited by many customers is that many ISPs have
> 576 MTUs set "inside" their networks and packets get fragmented.

I really don't buy that. Many or most backbone links have MTU >1500, and
MTUs <1500 outside of low-speed dialup connections aren't that common.
They are there, yes. But not that common.

My understanding of why a lower MTU is demonstratable better under Win95
is because the Win95 TCP stack is broken, and it is a good workaround.

Yep. And moreover; we mentioned the problem for the customers who use
WIN95 and try to get information from some WIN-NT servers, in case if
there is low-MTU (576) links somewhere between this client and server. I
suppose Win95 TCP/IP stack is not implemented correctly for this issue.

There are all sorts of people spouting all sorts of lies around Windows
newsgroups about why small MTUs are good; I think novice users are simply
getting drawn in by supposed experts.

In theory, small MUS and priority queuering can make delays less; on
practice (and if you remember about crasy and broken Win95 stack) the
best choise is to use MTU 1500 everywhere when some Win95 customers exist.

Aleksei Roudnev, Network Operations Center, Relcom, Moscow
(+7 095) 194-19-95 (Network Operations Center Hot Line),(+7 095) 239-10-10, N 13729 (pager)
(+7 095) 196-72-12 (Support), (+7 095) 194-33-28 (Fax)

How do I program my router to emulate one of those?

:slight_smile:

Cheers,-- jra

jra@scfn.thpl.lib.fl.us (Jay R. Ashworth) writes:

> Unless your ISP uses BBN Butterflies and C30 IMPs in its backbone, I
> would discount the odds of running into a link with an MTU of 576.

How do I program my router to emulate one of those?

You don't need to. It's already there. You just need to configure it:

  conf term
    int serial 0
    ip mtu 576
    no ip route-cache
    <repeat for all interfaces>
  no router bgp 109
  router egp 109
    ...
  ^Z

To simulate an IMP, you need to configure X.25 DCE interfaces. This is too
gross to discuss on a family mailing list.

:wink:

Tony

To simulate an IMP, you need to configure X.25 DCE interfaces. This is too
gross to discuss on a family mailing list.

Have you forgotten how to spell 1822? -s

To truly emulate the Butterfly's performance, wouldn't you have to do a
hardware mod and put a divide-by-64 on the clock?

:wink:

--zawada

Paul J. Zawada, RCDD | Senior Network Engineer
zawada@ncsa.uiuc.edu | National Center for Supercomputing Applications
+1 630 686 7825 | http://www.ncsa.uiuc.edu/People/zawada

>To simulate an IMP, you need to configure X.25 DCE interfaces. This is too
>gross to discuss on a family mailing list.

Have you forgotten how to spell 1822? -s

Fortunately, that little bit of hardware is no longer in production, or
supported. :wink:

Tony