It's Ethernet or Bust

The latest wireline data services revenue figures from the Big Seven U.S. telecom operators aren't pretty. Competition-driven price pressure, lackluster demand, and the migration of customers to more cost-effective Ethernet and advanced Internet Protocol (IP) services are all creating trouble for legacy private line, Asynchronous Transfer Mode (ATM), and Frame Relay sales. And because these legacy services make up the bulk of data revenues, they're putting a drag on the overall data services market.

And yet a sense of urgency seems strangely lacking among some executives at the major operators that Heavy Reading has profiled alongside other carriers in its new report, Carrier Ethernet Services: Who's Doing What. Third-quarter conference calls yielded barely a peep about Ethernet from executives at the Big Seven. Look in any of their most recent Securities and Exchange Commission (SEC) filings, and you won't find Ethernet-related services mentioned once. It's time those executives start paying attention to all the alarms going off around them and begin looking for ways to capitalize on the revenue opportunities for Ethernet that their own managers and others have told us they see in the market.

Set aside ramping DSL and Verizon Communications Inc.'s (NYSE: VZ) Enterprise Advance revenues, and you have some really bad news for the data market as a whole that is unlikely to get much better for quite a while. That's because the industry is accelerating through a painful transition from opex/capex-hungry data services toward opex/capex-friendly data services.

Consider these gloomy year-over-year figures from third-quarter 2004 financials:

  • AT&T Corp. (NYSE: T) – 10 percent drop in data services revenue.
  • MCI Inc. (Nasdaq: MCIP) – 16 percent drop in total data services revenue and a 7 percent drop in enterprise-specific data sales.
  • Sprint Corp. (NYSE: FON) – 8 percent drop in long-distance data services revenue, with private line and Frame Relay falling in the high single digits.
  • Qwest Communications International Inc. (NYSE: Q) – 4 percent drop in business data and Internet services revenue.
  • SBC Communications Inc. (NYSE: SBC) – 3 percent drop in high-capacity data services revenue.

Enterprise customers are clearly clamoring for the new stuff that gives them more bang for their buck and greater service flexibility. The story here is that carriers that depend on legacy data services can either be a victim of the transition or they can get out ahead of this mess. The sooner a carrier can psychologically and strategically make the shift to Ethernet and advanced IP/Multiprotocol Label Switching (MPLS) services – which cost less than comparable legacy services and are cheaper to deliver and support – the better off it will be.

With Ethernet services, big operators certainly have an advantage when it comes to fiber footprint – which is critical to delivering Ethernet. But as our new report demonstrates, they no longer have the luxury of slowly weaning their customers off legacy data services.

There are just too many players in North America (75 Ethernet service providers now, and counting) with an ever-growing list of intra-city and inter-city services (more than 330) that are making aggressive moves to supplant legacy services wherever they get the chance. And equipment vendors such as Atrica Inc., Cisco Systems Inc. (Nasdaq: CSCO), Fujitsu Ltd. (OTC: FJTSY; Tokyo: 6702), Riverstone Networks Inc. (OTC: RSTN.PK), and others are making it easier for even pint-sized operators to offer competitive Ethernet services with carrier-class features such as rapid restoration, end-to-end service-level agreements, seamless support for voice applications, robust service management, and greater scaleability than traditional Ethernet services.

No carrier that is heavily dependent on legacy service revenues is safe in such an environment. But in interviews conducted for this report, one RBOC spokesperson insisted that his company does not view Ethernet as strategic to its business. And an exec from a large IXC said his company was in no hurry to aggressively pitch Ethernet for fear of cannibalizing legacy services revenue. As painful as it may be, the big carriers have got to start shedding such inhibitions when it comes to marketing Ethernet, if they want to at least stabilize their overall data revenues, exclusive of DSL.

Meanwhile, Ethernet ranks at the top of the list of services – along with IP VPNs – that more than 200 enterprise users told Heavy Reading they plan to deploy in the next few years. The lack of noticeable outbound Ethernet-related comments and marketing from the Big Seven is even more puzzling, given that some 100 carrier professionals surveyed earlier this year by Heavy Reading said they expect Ethernet revenue growth to rival that of triple-play services in 2005.

Contrast the general lollygagging among the Big Seven to the gung-ho attitude of challengers such as Time Warner Telecom Inc. (Nasdaq: TWTC), which is unabashed in marketing its Ethernet story and aggressively targeting vulnerable RBOC markets. Lacking the trappings of legacy ATM and Frame Relay services in its portfolio, Time Warner and similar operators can fully embrace and advertise Ethernet's many advantages over legacy services – including cost-effectiveness, familiarity, ease of use, granularity, and scaleability.

While most of the Big Seven floundered in the third quarter, Time Warner reported a 23 percent year-over-year increase in its data and Internet services, fueled by Ethernet and IP-based sales. That should set off another round of alarms at the incumbents.

If you are looking for the overall data services market to grow, you're going to have to wait a while, because there's going to be a lot of churning under the surface before we see any rising tide. Carriers that insist on clinging to legacy product lines shouldn't be surprised when that lifeline turns into a dead weight.

— Stan Hubbard, Senior Analyst, Heavy Reading

Light Reading's Carrier-Class Ethernet in China roadshow will provide an invited audience of senior decision makers from service providers in China with a unique education in how to design and deploy profitable Ethernet services, employing original research written and presented by Heavy Reading analysts.

This event will take place in Shanghai on Nov. 30 and Beijing on Dec. 2. Those interested may register for it by
clicking here.

telecom_guru 12/5/2012 | 1:03:02 AM
re: It's Ethernet or Bust ILEC's too heavily invested in SONET, ATM and TDM. There is no viable threat yet (not like MSO's in Broadband, the only reason we now have DSL by the way). TWT you mention... let's get serious here, their access footprint cannot compare to ILEC's...just like all the IXC's in general. Just talk to them and they'll all say same thing, the bulk of the cost is in the access (they are at the mercy of the ILEC's fiber in the Access). Also Cisco, who is by far the most influential player in this space, is pushing a Layer3 routed solution and RFC2547 and bags MPLS (remeber MPLS agenda would mean less precious routers for them) they want a world where ILEC's are at their mercy... look at the ILEC deployments to date... its Cisco or Juniper in the IP nework but no major roll-out of Ethernet or MPLS...why? Cisco wants it that way!!! Oh...and he reguator excuse that has been conviniently used up until now... that a whole new discussion thread...

The rest of the world does not have this baggage that the US carrirs have (these seven carries spend almost all of the CapEx in NA). We need the MSO's to get serious about Ethernet Services but they are too focused on the consumer space so I would not hold my breath...Oh by the way...Cisco owns their IP network too.... stating to see a pattern here?
particle_man 12/5/2012 | 1:02:59 AM
re: It's Ethernet or Bust Maybe I missed your point, but RFC2547 is about MPLS/BGP VLANs. You seem to be saying that Cisco wants less MPLS in the network, but they are pushing MPLS VLANS (RFC2547)?

I'm sure that the router guys would prefer layer 3 VLANs to Layer 2 VLANs although Cisco has 70% of the Ethernet market too so they may win either way.
stanhubbard 12/5/2012 | 1:02:58 AM
re: It's Ethernet or Bust Some interesting comments. I just have two things to mention in response.

The RBOCS certainly have a big advantage on the fiber footprint, and my view is that they ought to make better use of that by adopting a more aggressive marketing stance with the Ethernet portfolios they have in place and the new services & features that they are looking at rolling out within the next 12-18 months. I think you'll see at least one major RBOC ramp things up in 2005, but I'd like to see more.

On the Cisco issue, I can see where people might think they are totally focused on high margin routers, but the fact is that the company is a big player with a demonstrated commitment to the carrier Ethernet service market. You don't have to look far to see Cisco's fingerprints all over the place. They have the 15454 MSPP supporting Ethernet services at Verizon, AT&T, BellSouth, Qwest, and elsewhere...the 6509 switch plus the 12000 routers in the MPLS backbone supporting Verizon's Transparent LAN service...7600 routers at SBC, etc. And they provide a lot of marketing support for numerous operators offering Ethernet servies.

Happy Thanksgiving
zolar 12/5/2012 | 1:02:39 AM
re: It's Ethernet or Bust Some very interesting posts...

1. Cisco makes more off their switches than they do routers, or any other product they have for that matter. As far as commitment to the carrier Ethernet service market, your dead wrong. I know this from direct experience. Cisco is very much a Enterprise business solution provider. It is over 80% of their revenue. They focus on the end user. Not the service providers. Don't care myself. I love Cisco anyway. But it's true.

2. The conversation regarding Sonet, ATM and TDM (TDM == Sonet BTW)...anybody with any type of infrastructure at all within the past few years has invested heavily in Sonet. Period. This was largely due to the fact that there really was no choice. Now that is changing more and more now. And there is plenty of choice...what good that does is very much worthy of debate as there is still no money in Telecom. It's pretty much a dead industry. I don't care who you work for or what platform your using. Some rides to the bottom are slower than others...but we are in fact, heading there. Too many players still. I'm on the boat myself, so if I have depressed any of you out there, your not alone. It's not so bad in sales, but if your in Engineering or Operations, or anything relating to anything Technical with Telecom, your wasting your time. Multiply that statement by a factor of 10 in regards to TWTC.

3. Stanhubbard...you make mention of the ONS 15454, and if anything, this little fella as it appears you know, makes the distinction between Ethernet and Sonet almost null and void...outside of the periodic little quirk. If anything, the ONS is the swiss army knife. Fibre Channel, TDM, Ethernet...you name it...it does it. That box revolutionized the industry.

dogmeat 12/5/2012 | 1:02:32 AM
re: It's Ethernet or Bust I want this type of service in US but it must be delivered over copper due to the lack of fiber coverage and the significant difficulty in installing fiber into my facilities.

I built out, with NTT, Ethernet to my buildings and eliminated any need for routers (bad for Cisco). NTT used ATM DSLAM's to land 25Mbps ATM and convert it to Ethernet. The backend is apparently MPLS (good for Cisco, though it's invisible to me).

The major problem in the US is the access gap from T-1 to DS3. I don't want to gang together a wad of T-1's to get what I need as far as raw access bandwidth. My retail locations need a cost effective combo of public/private bandwidth at around a 4Mbps data rate.

zolar 12/5/2012 | 1:02:12 AM
re: It's Ethernet or Bust Hi Dogmeat. It's interesting to hear Verio deploying such a legacy solution. Regardless, it works and sounds like it was cost effective for you.

The problem you describe regarding the access gap from T1 to DS3 really is no longer applicable. Fiber based facilites help...but are not a requirement. Be it wireless, or as you had indicated, bonding T1s, or doing a pure ethernet network, there are lots of options. NTT should be able to bond those T1s for you in a manner that is transparent to you completely. You just have a simple Ethernet port to plug into, so your option for being router free would still be option. Lots of options out there. Much more than a few years ago.
Sign In