[arin-ppml] debunking the myth that Moore's law helps
bicknell at ufp.org
Wed Dec 16 15:20:51 EST 2009
In a message written on Tue, Dec 15, 2009 at 08:52:18PM -0800, Michel Py wrote:
> This is just not true, and here's why: yes, Moore's law is still in
> effect. But guess what: link speed / bandwidth/ TE requirements / etc
> have evolved also, possibly even faster than Moore's law.
I'm not quite sure Moore's law is the right analogy, but I believe the
concept you put forth is sound. The same advances that drive link speed
faster (lower cost components, easier integration) also drive capability
in the high end boxes (able to route faster links). In some sense the
entire system moves in lockstep.
However, with high end routers there are actually a number of new
challenges. Much like PC's, they have hit the "Megahertz Wall" and
thus are moving to parallel processing solutions. The good news
is packet processing is often easy to parallelize, the bad news is
that it is all new code and hardware for the vendors. Someone is
going to have to pay for that development.
More interesting to me is the physics wall. Most long haul WDM
systems still have 10G channels. It turns out as you move past 10G
a number of properties of light cause added difficulty. There are
40G solutions, at a HUGE price premium over 10G solutions. The
innovation is in the 100G channel space, and requires developing
entirely new encoding standards.
I don't think these physics challenges are going to stop 100GE or
other technologies, but they are going to disrupt the curve. We've
gotten used to 10M ethernet replaced 5 years later by 100M for the
same cost, replaced 5 years later by 1000M for the same cost,
replaced 5 years later by 10000M for the same cost. It appears
rathern than seeing 100000M 5 years later and for the same cost,
we're going to see it 7 years later and for twice the cost.
(Ok, those are very crude estimates, but you get my drift.)
Someone else mentioned parallel links. There are ISP's today running
16x10GE, which is often the load balancing limit of the hardware.
They want to run 2x100G, or 4x100G, but those cards are not available.
They are still being tested in the lab. There's no ISP "afraid"
of running links in parallel, even where they could do 1x100G often
10x10G is cheaper right now due to the deployed long haul systems
Leo Bicknell - bicknell at ufp.org - CCIE 3440
PGP keys at http://www.ufp.org/~bicknell/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Size: 826 bytes
Desc: not available
More information about the ARIN-PPML