Any Hope for Ethernet over PDH?

One area that I strongly believe will see greater capex in 2008-2009 is Enterprise Access. (see “Enterprise Access Capex – A Ray of Hope?“) 

Cable modems forced the Telcos to dig DSL technology from the closet they were hiding it in order to remain competitive. The same forces are aligning today in enterprise access – but this time it’s dark fiber, PON, and even short range wireless that are the threats. What technology represents Telcos only hope of retaining business customers?

The quickest weapon the Telcos can bring to bear is Ethernet over PDH. It uses the venerable T1 and E1 line to deliver Ethernet services over the existing copper infrastructure.

EoPDH is a classic example of where a supplier ecosystem has developed in anticipation of large carrier deployments that haven’t materialized.

It’s possible that carriers will change their minds overnight, but the technology doesn’t address the two critical issues incumbent Telcos (who own the copper) will face in the coming Enterprise connectivity wars.

1. Insufficient bandwidth – While T1’s and E1’s can be bonded to deliver greater bandwidth at the end of the day it’s still 1.5Mb/s or 2Mb/s, data rates that haven’t kept with the times.
2. No Operational Cost Savings – There may be some savings associated with moving away from ML-PPP or Frame Relay but the amount of hardware in the field remains unchanged. FTTH brings substantial Opex in addition to higher speeds. EoPDH does not.

Technologies like hybrid fiber/VDSL, PON, or dark fiber Ethernet are the only real options that the Telcos have to defend their existing enterprise revenue streams. This is exactly what British Telecom concluded and is doing as part of 21CN. (see “BT’s 21CN – Reversing a Victorian Tradition“)

EoPDH is likely to remain a technology of last resort, used mostly in wireless applications more tolerant of slower pipes and potentially miles from any new infrastructure.

Corrections, Amplifications and differing opinions are welcome.

Update: In a rather odd co-incidence, Transwitch and T|Pack just announced cooperation and published a whitepaper on this very subject that is quite good.

————————————————

Breakdown of Markets and Players

EoPDH Markets:

• Enterprise – Use existing T1s and E1s to deliver Ethernet services to the existing enterprise. I estimate total volume in this market around 25k connections a year – virtually nothing compared with tens of millions of T1’s and E1’s deployed . Enterprise equipment suppliers need to offer EoPDH as part of their solution but carriers are not using it in material quantities today.
• Wireless – New 3G base stations are moving to Ethernet, particularly as wireless data proliferates. The only existing connection to remote cell sites is copper, so Ethernet over PDH is a potential solution. This is the nearer term and higher volume opportunity. Current volume unclear, readers are welcome to chime in.

EoPDH Component Suppliers:

• Enterprise – Adva (ADV.DE), Alcatel (ALU), Adtran (ADTN), RAD, Hatteras (pure play EoPDH company), Fujitsu
• Wireless – Nokia/Siemens (NOK), Ericsson (ERIC), Huawei, ZTE

EoPDH Component Suppliers:

• Transwitch (TXCC), Maxim (MXIM), Galazar Networks, PMC Sierra (PMCS), or developed in House; Nokia, Ericsson, and Adtran do this