Metro Buildouts

December 21, 2011 by Leave a comment

New metro construction continues at a rapid clip. Read through Verizon’s most recent 10-K and you’ll note that it publicly disclosed that it spent a material (at least material enough to disclose) amount of money on outside plant construction. Verizon specifically mentioned constructing new network to 9,000 cellular towers. Analyst reports suggest that AT&T and CenturyLink are also spending many hundreds of millions of capital to further expand their fiber optic networks. TW Telecom‘s investor materials suggest that it continues to add many hundreds, maybe even thousands, of new buildings to its vast fiber networks on an annual basis. In addition, Zayo’s press releases indicate that it is materially expanding its metro networks in a number of markets, including Minneapolis, Nashville, San Diego, Birmingham and Los Angeles to name a few.

So what’s my take on new network builds?

1.)   During the telecom boom in the late 1990’s, many telecom networks were built out.  These networks included both metro and long haul networks.

2.)    However, these build outs/networks only covered a fraction of the areas/buildings that, in 2011, ‘require fiber’.

3.)    The material amount of fiber optic construction that continues every year picks up more and more buildings that ‘require fiber’.

Filed under Dark Fiber 101 Tagged with , , , , , , , ,

More Long Haul Buildouts

December 12, 2011 by Leave a comment

Spread Networks isn’t the only entity building new long haul networks.  If ‘long haul’ is defined as ~100 route miles of new construction, then Zayo has also constructed several new long haul networks.  Three major Zayo builds include a multi-hundred–route-mile network connecting Indianapolis, IN to Evansville, IN to Louisville, KY,   a ~600-route-mile network throughout western Nebraska, and a long haul build between Toledo→Columbus, OH, which is part of a partnership.  Based on a cursory review of their website, it appears that Allied Fiber intends to construct major new sections of long haul network connecting Chicago→New York→Miami.  Several other companies have also constructed fresh ‘long haul’ networks in the northeast corridor (DC→NYC).  Fiberlight, for example, constructed network between Washington DC→Charlottesville, VA.  And new networks are being planned that will connect Southern VA→Atlantic Ocean.

Several recent analyst reports make reference to long haul as continuing to be ‘oversupplied’ or ‘underutilized’.  Perhaps?  But how does that jive with all the new ‘long haul’ construction?

Filed under Dark Fiber 101 Tagged with , , , , , , ,

A Greenfield Long Haul Network

December 5, 2011 by Leave a comment

Long haul networks are highly commoditized, and there is incredible competition within this space.  At least, that’s what conventional wisdom dictates. 

Within the last year, a new long haul fiber network was constructed.  Spread Networks quietly (at first) built and developed a long haul fiber network between Chicago and New York–despite the fact that numerous fiber optic networks already exist between New York and Chicago.  According to its website, Spread sells wavelengths and dark fiber across its network.  It appears, based on recent press releases that Spread has continued to extend and expand this network into several NYC and Northern New Jersey Data Centers.  Spread counts customers like Sidera and heavy hitting resellers like CFN as its customers/agents.

Spread Networks’ ownership/funding is unclear; however, given that James Barksdale is chairman of the board, I suspect that he and folks he knows are significant investors.  Spread is focused on a niche of the market, ultra-low latency trading, that didn’t exist in any substantive form 10 years ago.

Time will tell if Spread makes money for its shareholders. Yet, it is clear that new applications are driving the demand for new fiber networks.  These demands/niches didn’t exist 5-to-10 years ago and, most importantly, existing networks are not adequate to fulfill these demands.

But is Spread Networks a one-off?  Are they the only entity building a new long haul network?

Filed under Dark Fiber 101 Tagged with , , , , , , , , , ,

Back to Buildouts

November 7, 2011 by Leave a comment

I chatted about cell towers in my last blog post. Do you think the Apple Computer Data centers use twisted pair?  What about Google, Facebook, or salesforce.com’s data centers? Furthermore, do you think hospitals that are doing real time x-ray sharing or the Boeing facility that is communicating with the CIA on the drone project are using twisted pair to meet their bandwidth needs? …Of course not…

Have you heard of cloud computing (a virtual, off-site server)? Do you think employees use their office wifi to operate their mobile devices? How about managed video as a service (hosted video stored off- site)? Have you ever used WebEx or streamed music to your headphones at work? Do you think offices with more than 50 employees can get by on 3-6 megs of bandwidth? How about students?  There are ~100,000 K-12 schools and ~5,000 colleges in north America…I suspect they are thinking of creative ways to use a ton of bandwidth…

I think there are upwards of half a million commercial buildings/structures in the US that need fiber…maybe more…

And of course Google is stringing fiber optic cable to houses in Kansas City (with cash on hand of over $40B, Google certainly has the balance sheet to fund fiber buildouts). It seems Google has determined that twisted pair isn’t even sufficient  to meet the bandwidth needs of households in KC.

Bandwidth is growing…what might not be clear is the exact trigger-point for replacing legacy copper (twisted pair) with fiber. Technologies have been developed in recent years that allow ~50 megs of bandwidth to ride over twisted pairs of copper, but that is subject to quality issues/distance limitations.  Maybe, 100 megs = fiber is required, maybe 50 megs, or even as low as 10 megs; depending on the situation…

So, “what does this have to do with dark fiber?” Before I address this question, back to telecom network buildouts, including the first publicly announced major long haul buildout to hit the market since the telecom bust—Spread Networks…stay tuned….

Filed under Dark Fiber 101 Tagged with , , , , , , , , , , , , , , ,

A Decade of Growth

October 27, 2011 by 2 Comments

Funding dried up in the space circa 2001… many companies went bankrupt and were sold off in piece parts, others filed chapter 11 and re-emerged as strong well run companies.  2001 marked the end of an era of massive fiber buildouts…but that didn’t stop bandwidth from continuing its rapid growth…

 Something interesting happened between 2000 and 2011…no, house prices did not increase to $62M (they did increase to $316K in Nov 2007, then dropped back down to a current ~$250K).

In a turn of good fortune for telecom and bandwidth infrastructure providers, certain buildings that needed a single T-1 or a couple of T-1s in 2000, required many times more bandwidth in 2011. One obvious example is the cellular tower.  

First, thank you Steve Jobs.

As I remember it, in 2000, you got around 400 cell minutes for $60/month and you generally used your Nokia cell phone.  If you were an important business exec you even had a pager….  To support all of those calls a standard T-1 connection worked just fine for the vast majority of cell towers.

Do you have a 4G card, watch ESPN on your cell phone, use a 3G iPad, download music, log onto facebook, or stream real time stock quotes??  None of this was possible in the cellular world, much less existed (at least not in any meaningful fashion) back in 2000… Today all of these application exist, require bandwidth and are responsible for driving bandwidth growth on that lowly piece of steel known as the cell tower.

Is it any wonder that certain cell towers now consume far greater than 1.5 megs of bandwidth/second (the iPhone max download speed = 14.4 megs/second)?  If fiber is nearby the choice is simple… copper, yes copper (i.e., twisted pair) in 2000; fiber in 2011. 

 Oh, and there are estimated to be well north of 200,000 cell towers in the US….

 

Filed under Dark Fiber 101

The Telecom Bust

October 24, 2011 by Leave a comment

In 2000, global IP traffic was 75 petabytes/month.  Today that number is closer to 22,000 or greater than a 300 fold increase in eleven or so years.  To put that in perspective, the average price of an American house in 2000 was $207,000.  With a similar 300 fold increase that same average house would sell for a cool $62M…and you thought there was a housing bubble…therein lies the story.

Legacy telecom networks were largely built around a technology called ‘twisted pair’ invented by a gentleman named Alexander Graham Bell.  Ever heard of a ‘bell company’?  This technology and it’s many subsequent variants were deployed and widely used for the better part of 100 years.  Ever heard of a ‘T-1’…A T-1 supports ~1.54 megs of bandwidth/second. Guess how T-1 bandwidth gets from point A to point B? Yup, over twisted pair.  Need 3-6 megs of bandwidth? No problem, combine a couple of twisted pairs together and away you go.  If you need something north of 10 megs of bandwidth (and perhaps closer to 100 megs), the choice is simple:  fiber (assuming it’s nearby!)

Fiber optic cable was invented in the 1970s…by the early 1990s, amplification, a key requirement to use fiber economically over long distances, was invented.  In the early 1990s a phenomena hit the mainstream: AOL.  For approximately $20/month users could get access to a cool 12 kb/sec of bandwidth…in the same timeframe, one meg of IP bandwidth/second went, on the wholesale market, for upwards of $1,000.  Users demanded more cost effective bandwidth…thus, the race was on to construct fiber optic networks. 

Nearly all fiber optic network providers funded their network build outs by selling none other than dark fiber IRUs across most/all of their networks.  Many of these national networks  were built between 1996-2001 and connected to the same ~500 or so buildings nationwide.  In fact, in 2000 it could be argued that =a couple thousand buildings in the US really needed/used enough bandwidth to justify connecting fiber into the building…why not just add another twisted pair?  

Prices crashed when ten competitors showed up at many of the same key buildings, and, much like the story of Nacchio and Lay, dark fiber was (perhaps deservedly in this instance) the villain.  Many carriers who built out nationwide networks saw their fortunes diminish rapidly when their customers (who they sold dark fiber to) showed up at the same buildings and undercut their prices.  It’s a pretty simple micro-econ equation to figure out.  Let’s say two companies (or ten) invest hundreds of millions of dollars to build out networks that connect to (generally) the same buildings.  Assume the incremental cost to turn up another service at one of those buildings is virtually nil when compared to the cost of the initial investment; further assume shareholders who invested millions/billions of dollars expected to see a quick return on their money. 

The battle scars for those companies who survived the bust ran deep…I’m sure certain Sr. Execs thought:  If we just didn’t sell all that fiber to all of our competitors (and instead we simply raised additional debt or equity), we could have made a ton of money as pricing may have remained high for a long  time…they might have been right…we’ll never know…

The good news is that things were changing rapidly in the telecom space from 2001 – 2004…

Do you think it was a mistake for telecom providers to sell dark fiber from 1996 to 2001 (in order to fund their network buildouts?)

 

References:

http://en.wikipedia.org/wiki/Internet_traffic

http://en.wikipedia.org/wiki/Fiber-optic_communication

http://en.wikipedia.org/wiki/Twisted_pair

Filed under Dark Fiber 101 Tagged with , , , , , , , , ,

Lay and Nacchio

October 19, 2011 by 1 Comment

Ever heard of the term ‘telecom bust’?  Do you remember the names Ken Lay and Joe Nacchio?  They both contributed to the storied past of dark fiber.

 Lay and Nacchio

If you Google “Ken Lay Dark Fiber” or “Joe Nacchio Dark Fiber,” you’ll find a few interesting news articles at the top of your search results.  One of the search results that shows up is the SEC complaint filed against Nacchio.  Read through the first page and you’ll see that key to the complaint was the fact that Qwest relied on (dark fiber) IRU’s(1) to hit its recurring revenue targets.  The problem with this, according to the SEC, was twofold:  

1.)  IRUs result in one time (i.e., not recurring) revenue or, alternatively, amortized recurring revenue (i.e, revenue recognized over 20 years vs. one quarter) and

2.) These IRUs weren’t really needed… they were purported to be sham transactions where, for example, Enron (Lay) would buy unneeded capacity from a willing and able Nacchio and vice versa. 

The SEC alleged that, dark fiber IRUs  provided just the financial pop that Nacchio and Lay needed to meet their quarterly numbers.  I suspect this had nothing to do with dark fiber per se; rather, dark fiber was sold by the billions (perhaps even tens of billions) between 1996 and 2001.  There were ample, very legitimate, transactions done in the $100M+ range between 1996 and 2001… few, if any other telecom products were transacted at such high dollar amounts… so when Lay and Nacchio went looking for a solution to their business problems, they applied their business skills and looked for the low hanging fruit—dark fiber!

Much as I suspect is happening in the mortgage securitization business today, a dark shadow was cast over (very legitimate) dark fiber sales for some period of time.  Inside above board companies, I suspect questions were being asked: “can we sell big IRUs?”  “are they legal?”.  With criminal and civil complaints being filed, a wait and see attitude vis-a-vis dark fiber sales began to emerge…

But Lay and Nacchio were a secondary issue… a far larger issue was the telecom bust…

Do you think Lay and Nacchio gave a bad name to dark fiber?  Did they put a short term damper on the sales of dark fiber?  What was your experience?

Reference link: 

http://www.sec.gov/litigation/complaints/comp_nacchio19136.pdf

http://hsgac.senate.gov/012402partnoy.htm

(1)    An IRU is a legal interest that confers an indefeasible and exclusive right to utilize certain assets such as dark fiber.  Net/net, if you buy a dark fiber IRU, for all practical intents and purposes you own the fiber for the term of the IRU!

Filed under Dark Fiber 101 Tagged with , , , , , , , , ,

What is Dark Fiber?

October 18, 2011 by Leave a comment

The workhorse of the internet, cloud, data center, and 4G networks?  Fiber strands that are not lit?  An entire business for certain bandwidth infrastructure/telecom providers; a product not offered by other bandwidth infrastructure/telecom providers?  A ‘major’ extension of an enterprise LAN?   A core building block and efficient way to extend and expand telecom networks; a service that allows customers to manage their own networks?  A major target area of recent government stimulus funding?  Key to Google’s fiber to the home network in Kansas City or Verizon’s FIOS networks?  A ‘must have’ technology when building DAS networks?  A capital intensive business/product offering?   A product offering with a storied past in the telecom space?

Yes…..and more.

Dark fiber is rapidly becoming an optimal solution for businesses that have both the in house expertise to manage telecom equipment and need significant amounts of bandwidth,  generally > 1Gb/s-10Gb/s across multiple locations.

It’s interesting that dark fiber is becoming an optimal solution, given its storied past….

Filed under Dark Fiber 101 Tagged with , , , , , , , ,

Welcome to the Dark Fiber Blog

October 10, 2011 by Leave a comment

We are looking forward to sharing our thoughts and opinions with you regarding all things Dark Fiber.  We hope that along the way some of you will also provide us with your own thoughts.

Why are we doing this blog?  We are launching this blog so we can improve our engagement with the bandwidth infrastructure community.  Having a place where we can exchange information and opinions openly is important.

Our charter is to make sure that we keep the content fresh and relevant to our readers.

We will from time to time enter a post when we find articles on Dark Fiber or related topics of interest, as well as, provide our opinion on the product, industry, and it’s stakeholders.

Filed under Dark Fiber 101 Tagged with , ,