NAP Architecture


I had a lot to do with making the Pacific Bell ATM NAP work, so I'll weigh
in with a few comments while waiting for my plane out of Phoenix.

Q) Should I use a switch or a router?

A) If you are a disinterested third party L2 infrastructure provider, you
need to offer your NAP customers a L2 infrastructure. If you are an ISP and
you want to offer service to resellers, then you need to offer them a L3

The short answer is "you need a switch".

Q) What sort of switch should I use?

A) Use one that goes real fast on the interfaces, has a humongous aggregate
throughput, and offers plenty of buffer space for high bandwidth delay
product TCPs to bandwidth-hunt.

Early ATM switches didn't have sufficient buffer space for practical use of
TCP over UBR because Bellcore said they didn't need it. The DEC gigaswitch
suffers Head of Line blocking problems, the sort of mistake an engineer
makes who is building his first switch and doesn't like to read the

Some late model ATM switches have big buffers and work just fine for
Internet NAPs. Use UBR service. ABR service is unproven and may interact
with TCP congestion avoidance in bad ways, but this is unproven.

I should think a honking big frame switch (I prefer MAC frames) with big
buffers should work well, but do you want to be the first to try to run a
production NAP with one? I don't think you need fancy admission control,
rate control, explicit feedback congestion. Would be nice for the hardware
to have four to eight queues and some switches for finding the right header
bits to use to classify packets so that differential service might work
when ready.

A lot of folks think I'm an ATM bigot or at least suspect because I used to
talk to Bellheads, but a few years ago the only big switches that possibly
worked were ATM based. I still think that is true, because the old
gigaswitch doesn't work, due to HOL blocking. No one has tried anything
else yet, so there it stands. All I care is that it works and 1) simple 2)
big buffers does the trick today.

Of course, you could just sling an Ethernet hub in a rack and get started
that way. It's been done. :slight_smile:


Forgive my ignorance on these matters, but why haven't many NAPS tried
to be L1 based, or at least provide the option of private wire/fiber
between the larger customers in the same room. It seems to me that this
would significantly reduce the complexity and packet-loss we're currently
seeing. How long would it take to troubleshoot a cross-over FE compared
to trouble shooting two routers connected via a oversubscribed switch.
   Marketing types are concerned about how to bill and track these, but
there should be some easy ways around those issues.

--Ben Kirkpatrick
Data Products, Electric Lightwave, DID=360.816.3508
-not speaking for ELI, not even speaking-
"Consciousness: that annoying time between naps."

The PAIX is doing something very similar to this. The prices for
interconnects between cages are very reasonable.


The PAIX is doing something very similar to this. The prices for
interconnects between cages are very reasonable.

You are mistaken. In their original white paper (I still have it
somewhere) they proposed no cost cross connects between ISPs, but once
they were setup the business manager for PAIX set the fee at $1000 per
cross connect between ISPs after he calculated how much money he could
make by doing so. The thinking probably was "MFS does it, why not us?".

This was annoying because it destroyed one of the original features of the
NAP. However, you can't argue with a NAP facility that is completely sold
out (like PAIX).

Cross connects (fiber or copper) to carriers at PAIX are a much more
reasonable $75/month rate.


+------------------- H U R R I C A N E - E L E C T R I C -------------------+

While there is no vaild reason for a monthly reoccuring on a cross
connect. Unless, there is a tech going out and dusting the cross-connect
every month? Any carrier who charges monthly chrages on a wire hanging in
D-Ring, while collecting money for the cross connected parties to even be
in the same building, is not very nice.

A one time fee ($250 to $1,000) if they are doing the labor is much

One reason I like Telehouse so much.