[arin-ppml] A challenge to the assumption that a big DFZ is aproblem

Ted Mittelstaedt tedm at ipinc.net
Tue Dec 15 16:02:08 EST 2009

Michel Py wrote:
>> Ted Mittelstaedt wrote:
>> Today I can walk into the store and purchase a PC that has a CPU
>> in it that runs at a clock speed of at least 10 times of
>> most routers, and has at least 10 times the amount of ram, for
>> a quarter of the cost of the annual service contract for most
>> DFZ routers  (let alone the hardware cost)
> I'm sorry, but you have no idea how a core router actually works. I will
> try to make a comparison with a device that you may actually have seen.

It's not that I don't have any idea about how a core router works
internally, there's plenty of vendor documentation out there on current
designs if I wanted to start debating the merits of one or the other
of the current router designs.

It is that what your taking about is how the silicon is built TODAY.

What I'm saying is that it seems to me that there is no inherent barrier
in the silicon to coming out with routers that would do a much larger
DFZ  TOMORROW.  Granted, TODAY's designs might not scale up - but
is that a silicon problem or simply that more R&D is needed to find
better (or cheaper) router designs?

I will be the first to admit that this might have huge economic 
repercussions.  For example, the switchover of the US to digital TV
has had huge economic repercussions in the sale of consumer electronics,
since it's like half the showroom floor today is devoted to TV sales,
I'd assume that TV sales today are going great guns, driven by this.
Keeping in mind that most people probably WOULDN'T just go out and
replace a perfectly good TV for no reason.

But, is the economics of it our responsibility?  What is ARIN's
charter, folks?  Are you saying we should modify addressing policy
based on costs of hardware, or based on what it's POSSIBLE for
hardware to do?

I don't recall in the discussion of historical policy that the
reason that IPv4 allocations were originally limited was because
larger routers were expensive.  I believe the limitations were
because larger routers DIDN'T EXIST and there was no idea that
they would EVER exist.  Come on, folks, the most popular video
game back then was pong!!!!

It would be different if there was something in the silicon that
prevented this.  For example, you might have noticed recently that
the newer CPU designs don't seem to have higher and higher clock
speeds - that now, they are increasing the number of cores.  This is
because the laws of physics make it impossible to increase CPU speed
much more.  Thus, if we were still running on a MS-DOS world, then
power increases in the personal computer would have ended - since
MS-DOS didn't support more than 1 CPU.

> In my laptop bag, I carry this switch:

The thing about that is that your looking at a device which for the
last 20 or so odd years there's been competition to make them better
and faster at doing 1 thing - switching packets.  A PC is a general
purpose device, by definition it does nothing very well, but a lot
of things mediocre.  You could set your PC up as a sniffer - granted
not a great sniffer - or run a wordprocessor on it - you can't run
a wordprocessor on your switch.

That's not really what I was talking about.

> There is a reason why Cisco and Juniper still sell these multi-million
> dollar routers. They work (mostly). If someone had found a way to
> replace a CRS-1 or a T1600 with a PC, Cisco and Juniper would be out of
> business.

No they would not.  Didn't you know that all a Cisco PIX was is a PC?
It had a Intel CPU in it and all the PC support chips a PC of the time
had.  They'd adapt.

Nobody is going to come out with a PC that works as well at routing
packets as a purpose-built router, by definition you will always be
able to build a purpose-built hardware device that is faster than
a general-purpose device like a PC that's being used to solve a

BUT, that isn't what is needed, here.  What I'm really asking is, CAN a 
router be built that's fast enough for a much larger DFZ in the core? 
I'm not asking if such a beast would be "too expensive", I'm not asking 
if such a beast exists today, I'm asking if it's theoretically 
impossible to build such a device - in the same sense that it would be
impossible to build a square balloon out of thin, flexible rubber.


More information about the ARIN-PPML mailing list