From: jim bell <jimbell@pacifier.com>
Date: Thu, 16 May 1996 13:44:16 +0800
To: cypherpunks@toad.com
Subject: Re: Past one terabit/second on fiber
Message-ID: <199605151935.MAA15182@pacifier.com>
MIME-Version: 1.0
Content-Type: text/plain


At 10:44 AM 5/15/96 CDT, Daniel R. Oelke wrote:

>> "Wow", I said.  Far faster than the 2.5 Gb/sec transmission that is 
>> currently fairly standard for long-haul fiber trunks. 
>
>The ads say they are selling it - that doesn't mean shipping it... yet
>at least.  (Note that my employer is a direct competitor of Lucent 
>so I have a vested interest in setting the facts straight)

Yes, I should clarify that the article indicated that the whole system will 
be available in the "second quarter of 1996."  They say the fiber itself, 
"TrueWave" is available now.

>
>> I wasn't prepared, however, for page 38, in an article titled "Research 
>> Teams Achieve 1 Trillion bits a Second."  In fact, three separate groups 
did 
>> this.  I copy the article below.
>
>Yes - this is still very much a lab situation only though.
>It will be quite a few years before we hit that in real systems.

True.  In fact, it may be that we simply don't need that rate of 
transmission yet.  Maybe somebody familiar with routers can tell us what 
kind of CPU horsepower would be required to do that effectively.  Economics 
dictates that such an extraordinarily fast and multiplexed system would only 
be used when it is cheaper than the alternative, using multiple lower-speed 
fibers.  If anything, I think the main impediment to the use of the higher 
rates is fact that fiber signals must be broken down whereever the fiber is 
terminated, and the cost of that termination would be astronomical if it had 
to handle 1 Tb/s.  Chances are good that one of the few places that such a 
system would be economical would be long-haul undersea cables where 
breakouts are rare and terminations are few.

>> Even if we only consider that 20 Gb/second fiber from Lucent, that is 
>> equivalent to about 300,000 simultaneous voice calls.  
>
>An OC48 signal (~2.5 Gb/sec) will handle 48 T3's or 48 * 672 voice calls.
>Multiply by 8 for 20 Gb/second and you get 258048 voice calls.
>Pretty close to 300k I guess.

I was figuring they'd cut out silences...as well as echo-suppression 
cutouts.   Do they still do this?  Given modern fiber's capacity, I wonder 
if they bother.

>I don't know that any number fiber cable is "standard" but 
>36-fiber cable is not unusual.  To find the capacity of a cable, you
>have to cut the number of fibers in half (as you did) because
>generally each fiber is used only for a single direction of traffic.
>You then have to cut it in half again because phone companies have 
>everything redundant.  So, for a connection between two cities,
>there are generally 2 cables in different locations (so one backhoe
>doesn't get both), with on average only 1/2 the fibers in each carrying 
>paying traffic.

There's also a lot of 'dark fiber' out there, right?  Fiber laid as 
part of a cable but not activated, because it's not yet needed.  I got a 
look at a segment of the operation that laid a cable from Seattle (Vancouver 
BC?) to around San Francisco, about 5 years ago.  Got a chance to talk to an 
engineer, and ask a bunch of technical questions.  They laid three smaller 
(2" or so) plastic tubes in a larger outer tube, and the engineer said 
they'd later blow a cable through a single tube with compressed air.  He 
said they had no current plans to fill the other two tubes, because of lack 
of need.  And at the rate the fiber companies are improving transmission 
rates, it is unclear whether they will ever run out of capacity in such 
situations.


>> If we assume that the fiber cable costs $1/meter per fiber, and the cost of 
>> trenching, burial, and interconnects raise this to $10/meter/fiber, and if 
>> we generously assume that the average LD call goes 3000 miles (5,000,000m), 
>> that call occupies 1/150,000th of a $50 million fiber for a few minutes.  
If 
>> we suppose that the fiber has to gross $100,000,000 per year to pay for 
>> itself, and even if it's only operating at an average 10% load level(both 
>> assumptions are pessimistic, that only works out to a cost of 1.3 cents per 
>> minute per call.  That's why these LD phone companies are so scared:  If we 
>> can transmit Internet on fiber, that fiber can accept this extra traffic at 
>> very low marginal cost.
>
>I can't vouch much for your cost numbers - other than to apply the 
>factor of 2 adjustment noted above. 

My numbers are not totally a WAG (wild-ass guess) but they are probably out 
of date even if they had any resemblence to the truth a few years ago.  I'd 
appreciate hearing more accurate figures if you know them.  Don't divulge 
company secrets or anything; I'm only interested in ballpark figures for the 
industry as a whole.

> I would add that much of a 
>phone companies cost is in billing and customer service, etc.
>Not the cost of installing and maintaining the fiber and equipement.

This suggests that there would be a market for a LD phone company that 
charged, say, a yearly payment of $200-300 for essentially unlimited use.  (The main 
impediment to this would be regulatory; as I understand it LD companies have 
to pay local telco's for connections on a per-minute connect basis.  Is that 
right?  This needs to get fixed.)   Their billing costs would be very small. 


>Internet telephony should make the use of bandwidth even more
>efficient - thereby cutting costs.  The big guys who own
>the fibers will still make money - the pipes that carry internet
>traffic are still needed.  But the little guys will get squeezed out.
>(until they become ISPs ;-).  Internet traffic could theoretically
>be carried over this large amount of protection fiber (mentioned above)
>that is out there for a much lower marginal cost than the current 
>T3 or OC3 pipes that are being used.  The only "problem" being that these
>are lower priority channels, so that if a failure occurs anywhere,
>the traffic on them is dropped.  Most customers are demanding 
>high uptimes so much that the idea of a very lost cost, 
>but much less reliable service hasn't caught on yet.

I think the market will have to migrate towards such services.  A backup 
fiber is an asset whose capacity can be easily "mined" for a substantial 
payback.  True, its availability is not particularly reliable, but that's 
exactly what the Internet was intended to be able to use, right?  I suspect 
that the real reason Internet hasn't migrated to such transmission systems 
is simply that its needs still don't demand anywhere close to the 
current-available fiber technology.  However, if traffic is tripling per 
year we'll need to see a substantial migration to full-fiber links within 
3-5 years.

Jim Bell
jimbell@pacifier.com