Article Info

The Real Carrier Ethernet Opportunity

It is amazing how little can change in three years. I spent a great deal of time working on Carrier Ethernet in 2004 and 2005, and the presentations I saw at the Lightreading Ethernet Conference and Expo were no different than the ones I saw in 2004.

Equipment makers such as Ciena (CIEN) sang the praises of Carrier Ethernet (all true) and spoke of the various impediments to deploying it: standardization of inter carrier interfaces, administration & operation, quality of service.  It strikes me that the bigger problem is much more basic than the ones being presented.

Several times during the conference carriers (the big ones in particular)lamented their inability to acquire rights to use the fiber that reaches customers. Enterprise customers want Ethernet services. Big service providers like Verizon (VZ) and AT&T (T ) want to sell them services. But the issue is that the fiber to enable the connection between the big service providers with global reach and the big customers looking for 10GbE and 1GbE connections either doesn’t exist, or it sits in the hands of others who can’t provide global scale.

It’s a layer zero problem. In the US, only 12% of Enterprises have fiber connectivity. Only 20% of North American cell sites have fiber connectivity. While Ethernet over Copper is a neat technology and provides incremental benefits, it does not provide enough incremental bandwidth to users and opex savings to carriers to drive a major capex cycle. ML-PPP and Frame Relay are ‘good enough’ in comparison.

This reinforced my belief that local fiber is a scarce asset, particularly in situations where only one company owns an optical right-of-way. People are aware of companies that own fiber in major metropolitan areas such as Level3 (LVLT), Cogent (CGNT) or Abovenet (, but in many cases these companies all own duplicate connections to the same buildings.

Zayo Bandwidth is a consolidator of fiber carriers in second tier markets. While on the surface this doesn’t appear interesting, a sole-source connection in Gary, Indiana could be more valuable than a triplicate fiber connection inside the Loop in Chicago. The value in fiber carriers is in the uniqueness of their connections, not their quantity. This is the asset that provides the greatest sustainable competitive advantage.

Therefore, the real opportunity to benefit from Carrier Ethernet is not in chips, optics, or equipment – the opportunity is in the new class of carriers that are trenching unique fiber, driving penetration from 12% to 20% to 40%. Find companies that own this fiber or are trenching this fiber and you will find the best way to benefit from the roll out of Carrier Ethernet services out to the enterprise.

Disclosure: Author owns shares of Abovenet


Comments are disallowed for this post.

  1. Andrew, one such carrier is 4Connections, which is headquartered in Elizabeth, NJ.

    From Fiber Optics Online

    4Connections Adds More Lit Services To Its Existing Dark Fiber Network

    Parsippany, NJ – 4Connections, a New Jersey and New York metro lit and dark fiber provider, announces recently it will increase its “lit fiber” services to its existing “dark fiber” network, enabling the company to provide service and choice to end-users in remote areas, where often there is only one carrier available. Cont. at:

    Posted by Frank A. Coluccio | October 17, 2007, 10:10 AM
  2. * Interesting … It’s like “Spy vs. Spy” figuring this stuff out.
    * Over 5 Years ago the lament went something like “95% of Enterprises are within a kilometer of fiber while less than 5% have a fiber connection” ( so 12% seems like a lot to me )
    * Yet, it seems like the majority of “big customers” have fiber already, or have a way to get it ( via Co-Location Datacenters )

    Posted by iain verigin | October 17, 2007, 4:06 PM