Is Google Fiber a model for Municipal Networks?

Mailman List Mirror (Read Only)
1

I've been searching for a few days on information about Google
Fiber's Kansas City deployment. While I wouldn't call Google
secretive in this particular case, they haven't been very outgoing
on some of the technologies. Based on the equipment they have deployed
there is speculation they are doing both GPON and active thernet
(point2point).

I found this presentation:
http://static.googleusercontent.com/external_content/untrusted_dlcp/research.google.com/en/us/pubs/archive/36936.pdf

It has a very good summary of the tradeoffs we've been discussing
regarding home run fibers with active ethernet compared with GPON,
including costs of the eletronics compared to trenching, the space
required in the CO, and many of the other issues we've touched on
so far.

Here's an article with some economics from several different
deployments: http://fastnetnews.com/fiber-news/175-d/4835-fiber-economics-quick-and-dirty

Looks like $500-$700 in capex per residence is the current gold
standard. Note that the major factor is the take rate; if there are two
providers doing FTTH they are both going to max at about a 50% take
rate. By having one provider, a 70-80% take rate can be driven.

Even with us a 4%, 10 year government bond, a muni network could finance
out a $700/prem build for $7.09 per month! Add in some overhead and
there's no reason a muni-network couldn't lease FTTH on a cost recovery
bases to all takers for $10-$12 a month (no Internet or other services
included).

Anyone know of more info about the Google Fiber deployment?

2

I've been searching for a few days on information about Google
Fiber's Kansas City deployment. While I wouldn't call Google
secretive in this particular case, they haven't been very outgoing
on some of the technologies. Based on the equipment they have deployed
there is speculation they are doing both GPON and active thernet
(point2point).

I found this presentation:

http://static.googleusercontent.com/external_content/untrusted_dlcp/research.google.com/en/us/pubs/archive/36936.pdf

Its active ethernet. They looked at PON but ultimately rejected it since

it fell below their speed goals (can't do gig connections on any flavor of
PON today).

Here is the architecture document:

It has a very good summary of the tradeoffs we've been discussing
regarding home run fibers with active ethernet compared with GPON,
including costs of the eletronics compared to trenching, the space
required in the CO, and many of the other issues we've touched on
so far.

Here's an article with some economics from several different
deployments:
http://fastnetnews.com/fiber-news/175-d/4835-fiber-economics-quick-and-dirty

Looks like $500-$700 in capex per residence is the current gold
standard. Note that the major factor is the take rate; if there are two
providers doing FTTH they are both going to max at about a 50% take
rate. By having one provider, a 70-80% take rate can be driven.

Even with us a 4%, 10 year government bond, a muni network could finance
out a $700/prem build for $7.09 per month! Add in some overhead and
there's no reason a muni-network couldn't lease FTTH on a cost recovery
bases to all takers for $10-$12 a month (no Internet or other services
included).

Anyone know of more info about the Google Fiber deployment?

The biggest factor that Google has going for them is they are their own
gear manufacturer, both the in home stuff and the access network. They
invited several manufacturers to test but then sent them all packing. They
are doing a ring (actually several rings) of Ethernet with nodes that then
connect down to the neighborhood level.

3

From: "Leo Bicknell" <bicknell@ufp.org>

Here's an article with some economics from several different
deployments:
http://fastnetnews.com/fiber-news/175-d/4835-fiber-economics-quick-and-dirty

Looks like $500-$700 in capex per residence is the current gold
standard. Note that the major factor is the take rate; if there are
two providers doing FTTH they are both going to max at about a 50% take
rate. By having one provider, a 70-80% take rate can be driven.

I was seeing 700 to drop, and another 650 to hook up; is that 5-700 supposed
to include an ONT?

Even with us a 4%, 10 year government bond, a muni network could finance
out a $700/prem build for $7.09 per month! Add in some overhead and
there's no reason a muni-network couldn't lease FTTH on a cost recovery
bases to all takers for $10-$12 a month (no Internet or other services
included).

This is the class of numbers I was hoping to get to, yeah.

Cheers,
-- jra

4

In a message written on Sun, Feb 03, 2013 at 05:03:52PM -0500, Jay Ashworth wrote:

> From: "Leo Bicknell" <bicknell@ufp.org>
> Looks like $500-$700 in capex per residence is the current gold
> standard. Note that the major factor is the take rate; if there are
> two providers doing FTTH they are both going to max at about a 50% take
> rate. By having one provider, a 70-80% take rate can be driven.

I was seeing 700 to drop, and another 650 to hook up; is that 5-700 supposed
to include an ONT?

I believe the $500-$700 would include an ONT, if required, but
nothing beyond that. "Hook up" would include installing a home
gateway, testing, setting up WiFi, installing TV boxes, etc.

So in the model I advocate, the muni-network would have $500-$700/home
to get fiber into the prem, and the L3-L7 service provider would
truck roll a guy and supply the equipment that comprise another
$500-$700 to turn up the customer.

In Google Fiber's model they are both, so it's probably $1000-$1400
a home inclusive. $1400 @4% for 10 years is $14.17 a month per
house passed.

5

Sure, Verizon has been able to get their cost per home passed down to $700
(http://www.isuppli.com/Home-and-Consumer-Electronics/MarketWatch/Pages/Veri
zons-FTTH-Expansion-Stoppage-Takes-Many-by-Surprise.aspx), but that does not
include the drop, ONT, nor any home wiring to get from the ONT to the CPE
within the home. That's still another $650 or so (slide 11 of
http://www.natoa.org/events/NATOAPresentationCalix.pdf and
http://www.nytimes.com/2008/08/19/technology/19fios.html?pagewanted=all&_r=0
). That's down from $1,021 in 2005 and $850 at the end of 2006
(http://connectedplanetonline.com/home/news/verizon_touts_fios_092706/).

While my $DAYJOB's access vendor has a range of outdoor ONTs with different
features, they generally cost from $400 to $700 apiece (plus ONT housing,
power supply, battery, and power cable). If we assume Verizon, because of
their purchasing power, can negotiate a fantastic discount, their average
ONT cost probably still exceeds $200 apiece. Google is not concerned with
traditional POTS in their offering, so they don't have to worry about backup
power requirements (and costs), plus they're doing ActiveE, not GPON, so
despite their low volume, Google probably has ONT costs somewhat similar or
less than Verizon.

Real-world FTTH complete overbuilds among RLECs (rural incumbent LECs) are
typically between $2,000 and $5,000 per home served (that includes the ONT
and customer turn-up). Slide 13 of
http://www.natoa.org/events/NATOAPresentationCalix.pdf shows an average of
$2,377 per home passed (100% take rate). You can see on Slide 14 how the
lower households per square mile leads to substantially greater costs.

Looking again at the Verizon FiOS build, where there may be a complete
overbuild but not every copper customer is converted to FiOS, the cost per
home passed does not equal cost per home served. Note this BusinessWeek
quote regarding FiOS, "He estimates the project will end up having cost
Verizon $4,000 per connected home."
(http://www.businessweek.com/magazine/content/11_13/b4221046109606.htm)
And for Verizon's cost per home passed: "Consider the total project cost of
Verizon's FiOS, $23B, and then divide that not by the 17M homes passed (as I
did), but with the actual subscribers (5,1M), This would result in a cost
per subscriber of $23B/5.1M = $4,500."
(http://seekingalpha.com/article/778871-challenges-and-opportunities-for-goo
gle-s-fiber-project-a-reality-and-sanity-check). From the same Seeking
Alpha article, FiOS' cumulative historical cost per home passed is $1,352.

Remember that Google cherry-picked which city it would serve, so it was able
to identify location that is likely less challenging and expensive to serve
than the average. A lot of Google's Kansas City build will not be buried
but hung on power poles, and they're starting their builds in areas of the
city where there was a significant level of customer interest. All of that
is going to translate into a lower cost per customer served than a service
provider who decides to overbuild an entire locale, regardless of customer
interest. So while Google may be able to pull off a $1,400 expenditure per
home passed, Jay can't use that price point in his own calculation unless
he's in similar construction environment and takes Google's selective
deployment approach.

Frank

6

When comparing costs of building (per home passed/connected), it is also
important to see if those quoted costs include the regulatory costs of
dealing with cities.

If a municipal project won't suffer costs of negotiating for
diggging/building permits, already has the land to build the "CO", and
won't be delayed by stalled paperwork, this could represent a
significant difference compared to an incumbent that constant hits a
brick wall of bureaucracy which cost money and delays the project. (In
Canada, a delay of a couple of months can cause a delay of 1 year due to
winter).

In the case of Google, with cities begging on their knees to get
Google's project, I suspect they didn't have many problems getting the
city's cooperation. Yet, there have been stories of delays due to
bureaucracy.

In the case of Verizon, I suspect those bureaucracy costs are much higher.

The other aspect which you need to compare is existing infrastructure.
If there are already conduits between CO and neighbourhood, and there is
room inside, you can just blow your new fibre through them which costs a
lot less than having to dig and install new conduits.

(and "space available" is one of the issues that lead telcos to go GPON
instead of wanting 1:1 strand to home ratios since that requires much
more conduit capacity the closer you get to your point of aggregation.

So when comparing both Verizon and Google projects, you need to factor
exactly how much needed to be built for them, and how much will be
needed for you. If you have 0 existing conduits and need to build 100%
of your FTTH plant, while Verizon had x% of conduits already built, then
your cost per home may be higher.

BTW, out of curiosity, how many spare copper pairs were traditionally
provisionsed on a cable run that passed 100 homes ? And in a fibre
system, do you keep the same ratio of extra strands for each home passed ?

7

Nice get; that will make very interesting reading today. Thanks.
-- jra

8

Scott Helms wrote:

Here is the architecture document:
http://static.googleusercontent.com/external_content/untrusted_dlcp/research.google.com/en/us/pubs/archive/36936.pdf

The document, seemingly, does not address drop cable cost
difference.

It does not address L1 unbundling with WDM-PON, which
requires fiber patch panel identical to that required
for SS, either.

As for power consumption at CO, all the transmitters do not
have to have power consuming LDs but can just have modulators
to modulate light from a shared light source, which has already
happened with QSFP+:

  Optical Transceivers and Coherent Optics - Cisco

  How do you generate light in silicon?

  Actually, we don't. Silicon is a bad material to try and
  build lasers in. Some silicon lasers have been demonstrated,
  but these are completely impractical. As it turns out there's
  no need to build a silicon laser: lasers are already very
  inexpensive (remember, there's already one in every PC
  - inside the CD/DVD player). The challenge has been finding
  an inexpensive way to attach the lasers to silicon. Solving
  this problem, and the related one of inexpensively attaching
  optical fibers to silicon, is a key piece of Luxtera's
  intellectual property. We think of a laser as being just
  like a DC power supply – only it provides a steady stream of
  photons rather than electrons.

            Masataka Ohta

9

In a message written on Sun, Feb 03, 2013 at 09:53:50PM -0600, Frank Bulk wrote:

Sure, Verizon has been able to get their cost per home passed down to $700

To be fair, Verizon has chosen to build their FIOS network in many
expensive to build locations, because that's where they believe
there to be the most high profit customers. While perhaps not the most
expensive builds possible, I would expect Verizon's FIOS experience to
be on the upper end of the cost scale.

Real-world FTTH complete overbuilds among RLECs (rural incumbent LECs) are
typically between $2,000 and $5,000 per home served (that includes the ONT
and customer turn-up). Slide 13 of
http://www.natoa.org/events/NATOAPresentationCalix.pdf shows an average of
$2,377 per home passed (100% take rate). You can see on Slide 14 how the
lower households per square mile leads to substantially greater costs.

Rural deployments present an entirely different problem of geography. I
suspect the dark fiber model I advocate for is appropriate for 80% of
the population from large cities to small towns; but for the 20% in
truely rural areas it doesn't work and there is no cheap option as far
as I can tell.

And for Verizon's cost per home passed: "Consider the total project cost of
Verizon's FiOS, $23B, and then divide that not by the 17M homes passed (as I
did), but with the actual subscribers (5,1M), This would result in a cost
per subscriber of $23B/5.1M = $4,500."

But Verizon knows that take rate will go up over time. Going from
a 5.1M -> 10M take rate would cut that number in half, going to the
full 17M would cut it by 70%. Fiber to the home is a long term play,
paybacks in 10-20 year timeframes. I'm sure wall-street doesn't want to
hear that, but it's the truth.

Remember that Google cherry-picked which city it would serve, so it was able
to identify location that is likely less challenging and expensive to serve
than the average. A lot of Google's Kansas City build will not be buried

True, but I think it means we've bound the problem. It appears to
take $1400-$4500 to deploy fiber to the home in urban and suburban
areas, depending on all the fun local factors that effect costs.

Again, if the ROI calculation is done on a realistic for infrastructure
10-20 year time line, that's actually very small money per home. If
it's done on a 3 year, wall street turnaround it will never happen as
it's not profitable.

Which is a big part of why I want municipalities to finance it on 10-30
year government bonds, rather than try and have BigTelco and BigCableCo
raise capital on wall street to do the job.

10

And look what appeared in my mailbox just now:

http://broadcastengineering.com/ip-network/google-s-high-speed-fiber-installation-provides-economic-growth-kc

Cheers,
-- jra

11

*sigh*

I'd gladly pay $5000 NRC to get fiber to my house. I only wish it
were that simple. :frowning: Heck, if they wanted longer-term ROI, I'd pay
$5000 NRC and $200 MRC for a fiber connection to my house, if
only there were someone who could provide it. I suspect the real
costs are much higher, and that's why there's nobody willing to do
it for that cheap.

Matt

12

No, Matt; in a sufficiently dense deployment, it appears you can actually
get it done for that money, based on actual deployment results.

If my project pans out, I'll give it to you for less than that. :slight_smile:

Cheers,
-- jra

13

I think the problem with your model is that the same one
Google faced; you don't divide your cost based on the number
of homes connected, you divide by the number of people forking
out money for it.

Building infrastructure to 10,000 homes doesn't work if 9,999
of them say "no thanks, I'm happy with my current cable TV";
that last person's gonna have a heck of a bill, or you're going
to go bankrupt subsidizing them.

Google Fiber's "sign up, and if we get enough signups, then
we'll build" model seems to be the only sane way to ensure
that you won't be left holding the bag if not enough subscribers
opt in to the service to fund it.

Now, if only we had a system for signing up to show our
support for a build like that here in the bay area... ^_^;

Matt

14

Scott Helms wrote:

Here is the architecture document:

http://static.**googleusercontent.com/**external_content/untrusted_**
dlcp/research.google.com/en/**us/pubs/archive/36936.pdf<http://static.googleusercontent.com/external_content/untrusted_dlcp/research.google.com/en/us/pubs/archive/36936.pdf&gt;

The document, seemingly, does not address drop cable cost
difference.

It does not address L1 unbundling with WDM-PON, which
requires fiber patch panel identical to that required
for SS, either.

They're not doing WDM-PON or any flavor of PON at all. Its entirely an
Active Ethernet deployment.

As for power consumption at CO, all the transmitters do not
have to have power consuming LDs but can just have modulators
to modulate light from a shared light source, which has already
happened with QSFP+:

        Optical Transceivers and Coherent Optics - Cisco

        How do you generate light in silicon?

        Actually, we don't. Silicon is a bad material to try and
        build lasers in. Some silicon lasers have been demonstrated,
        but these are completely impractical. As it turns out there's
        no need to build a silicon laser: lasers are already very
        inexpensive (remember, there's already one in every PC
        - inside the CD/DVD player). The challenge has been finding
        an inexpensive way to attach the lasers to silicon. Solving
        this problem, and the related one of inexpensively attaching
        optical fibers to silicon, is a key piece of Luxtera's
        intellectual property. We think of a laser as being just
        like a DC power supply – only it provides a steady stream of
        photons rather than electrons.

Masataka, are your trying to participate in the conversation or sell gear?
The laser used in your DVD player is NOT suitable for a broadband
deployment.

15

Rural deployments present an entirely different problem of geography. I
suspect the dark fiber model I advocate for is appropriate for 80% of
the population from large cities to small towns; but for the 20% in
truely rural areas it doesn't work and there is no cheap option as far
as I can tell.

Why do you want a muni to put in fiber but not light it? Wouldn't it make
more sense to simply put in fiber runs and let company's lease space?
Trenches don't really degrade over time and there is a lot less of a
requirement for cooperative troubleshooting and far less blame game.

Which is a big part of why I want municipalities to finance it on 10-30
year government bonds, rather than try and have BigTelco and BigCableCo
raise capital on wall street to do the job.

I certainly sympathize with wanting independent connections but most cities
have their own budget concerns and doing a bond on a fiber network they
can't or don't light is a harder pay back on one that they do light. I'd
suggest either layer 2 sharing (ethernet with per sub VLANs) or trench
sharing as above.

16

Scott Helms wrote:

The document, seemingly, does not address drop cable cost
difference.

It does not address L1 unbundling with WDM-PON, which
requires fiber patch panel identical to that required
for SS, either.

They're not doing WDM-PON or any flavor of PON at all. Its entirely an
Active Ethernet deployment.

My point is that their comparison between SS and PON is insufficient.

As for power consumption at CO, all the transmitters do not
have to have power consuming LDs but can just have modulators
to modulate light from a shared light source, which has already
happened with QSFP+:

Masataka, are your trying to participate in the conversation or sell gear?

My point is that form factor reduction by silicon photonics
excludes LDs.

The laser used in your DVD player is NOT suitable for a broadband
deployment.

Do you understand that QSFP+ is for 10G Ethernet?

One or two (or three, maybe) shared light source in CO can
have much better quality, which can be distributed to all
the transmitters using splitters and EDFA, which does not
consume a lot of power.

            Masataka Ohta