Huge Uptake in in Business Ethernet Services led by Competitive Carriers


Business customers are buying Ethernet services from legacy and competitive carriers as well as from MSOs.  Cost effective performance, simplicity of operations, scalability and the ability to use the same access technology as in the enterprise LAN are somme of the reasons for this uptake.  In addition to the extra bandwidth and the variety of services Ethernet supports, small- and medium-sized businesses appreciate Ethernet’s flexibility and lower cost.  Carriers and their customers like the fact that Ethernet adapts to their premises based (switch/router) equipment, which helps them to grow their businesses quickly. Wholesalers are doing the same things as their retail counterparts

Why Business Ethernet?

The top three reasons carriers are deploying Business Ethernet are to transport mobile backhaul traffic, to meet the needs of high-end enterprises and to serve businesses with multiple locations.  For many years, Frame Relay service was used for the latter, but being almost 20 years old, Frame Relay has been replaced by “Metro Ethernet” which also goes by the monickers “Carrier Ethernet” and “Business Ethernet.”  Note that Carrier Ethernet is also used by many carriers (e.g. AT&T U-Verse) to transparently deliver IP-TV services to residential customers.  However, that is not the focus of this article.

Ethernet over Fiber or Copper depends on Bandwidth Requirements

To truly realize the potential and power of Metro Ethernet, carriers need to build-out fiber infrastructures closer and closer to their customers.  Optical Ethernet from the customer premises to the carrier’s POP is the best, because it facilitates “liquid bandwidth.”  That means that a business customer can automatically request his Ethernet speed to be increased or decreased without having to add or remove a physical line.  The addition of a new facility involves a truck roll and may take significant time.  This is not necessary with FTTP because of the much higher access bandwidth of fiber, when compared to a repeatered 4 wire copper circuit (e.g. T1/DS1 or T3/DS3) or bonded (2 wire) DSL.

When fiber to commercial buildings is not available or possible,  the fiber is terminated in a network node close to business customers and then extended by copper tail circuits -either bonded DSL or n x T1/DS1.  The Metro Ethernet sweet spot today is somewhere between 10M b/sec and 100 M b/sec- much lower than what many pundits were predicting for 2000 and 2001 – just before the telecom crash.  Those speeds can effectively be met with Ethernet over Copper, which provides reasonably good scalability from 3 – 20 M b/sec.

According to Erin Dunne, Director of Research Services for Vertical Systems Group, just over half of U.S. business Ethernet installations are delivered over Direct Fiber.  Ms Dunne states that “SONET/DWDM is the second most prevalent Ethernet access technology.  Other technology alternatives include TDM, bonded copper, Coax/HFC, FWL and others.  While fiber is the access technology of choice for Ethernet service delivery, alternatives to fill gaps in fiber coverage will be required for many years.  This is good news for vendors of equipment that addresses the flexible access requirements service providers are demanding.”

Comment and Analysis

What’s so amazing to me is that it took over 8 years for the Business Ethernet/ Metro Ethernet market to take off. The IEEE 802.3 Ethernet First Mile Standard was completed in 2003, many MEF specs were solid at that time and the ITU-T had started a massive effort to standardize end-to-end “Carrier Ethernet,” including OA&M as well as Performance Monitoring and Protection Switching.

Many “new age” Ethernet carriers and start up Ethernet network equipment vendors went bankrupt or were acquired for a fraction of what VC investors thought they were worth.

Now – 8 years later- the volume of Ethernet bandwidth purchased by enterprises in the U.S. has surpassed the aggregate bandwidth for legacy circuits, according to new research from Vertical Systems Group.

2011 marks the tipping point of a surge in the installation of Ethernet connections. Looking ahead to 2015, Ethernet bandwidth is projected to more than double based on Vertical’s latest analysis of enterprise requirements.

“Boosted by a 10x surge in the past five years, Ethernet bandwidth has overtaken legacy bandwidth in the U.S. market,” said Erin Dunne, director of research services at Vertical Systems Group. “This milestone fittingly coincides with the ten year anniversary of the MEF – an organization that has successfully fostered the deployment of carrier-class Ethernet services throughout the world.”

XO Communications Business Ethernet Services

A survivor of the dot com bust and telecom crash of 2001-2002, XO Communications is quietly making a name for itself in delivering Ethernet services to business customers.  While AT&T, Verizon, and TW Telecom are still the U.S. leaders, according to Vertical Systems Group Mid Year 2011 Leaderboard, XO is in 7th place and moving up fast.  The company sells voice and data services to SMBs and also offers wholesale facilities to other carriers/re-sellers.

The Ethernet services XO offers to business customers include:

  • Ethernet Private Line – Point-to-point Ethernet connectivity ideal for businesses looking for dedicated bandwidth between office locations
  • Ethernet Hub (previously called Ethernet Virtual Private Line)– Point-to-multipoint Ethernet solution ideal for connecting branches or offices to centralized headquarters or data center
  • Ethernet VPLS (previously called Ethernet Private LAN) – Multipoint-to-multipoint Ethernet WAN ideal for enterprises to connect key locations, transport special applications with protocol transparency, and maintain separation of different network domains
  • Ethernet Access (to an IP WAN) – affordable, scalable Ethernet access to deliver Dedicated Internet Access, VoIP and MPLS IP-VPN services across your enterprise

A complete description of XOs Ethernet Services is described in their  Ethernet Services Overview document.  The company also provides an informative podcast describing why business customers are interested in Ethernet services and how XO meets that need.  To watch that video, please visit:

According to Don MacNeil, vice president of Carrier Services for XO Communications, Ethernet has been key to the company’s strong growth since its founding in 1996, a trend he foresees continuing in the future.

The XO network can provide Ethernet services to over 10 million U.S. businesses in the metro areas the company serves. This represents the majority of all U.S. businesses and the XO network is present in all top 25 metro markets. XO’s Ethernet-over copper services are now available in 442 central offices in 40 metro markets reaching 1.5 million U.S. businesses.

For more information on XO’s Business Ethernet offerings as well as other XO telecom services for SMBs, please contact:  Michael E. Weiss,  XO Communications Major Accounts at: or phone:  510-580-6380.  Mr Weiss attended the Oct 12th IEEE ComSocSCV meeting on cloud networking and talked with attendees during our networking reception.  He also attended our very successful social networking dinner in August.

For more information about ComSocSCV please visit:, ComSocSCV LInked IN group, Facebook page and the Sept 2010 and Jan 2011 editions of IEEE Global Communications Newsletter (GCN).  There will be another article about ComSocSCV in the Nov 2011 GCN

Intel CEO Paul Otellini talks up Innovation at SCU President Speakers Series

“Create and extend computing technology to connect and enrich the lives of every person on earth.” According to Intel, it is the mission and kind of thinking that has defined and shaped the comapny’s actions for the past four decades.  Intel’ s world class semiconductor processing technology, enables the company to design and build microprocessors that serve as the foundation for the world’s computing devices.

Intel is dedicated to solving some very complex problems: How do we push transistor technology beyond the seemingly immovable laws of physics? How can we empower people in the remotest villages around the world with tools of the digital age? How will we prepare young people, so they will continue to innovate and solve some of our world’s biggest challenges?

Paul Otellini confronts these issues everyday as President and Chief Executive Officer of Intel Corporation. After being named CEO in 2005, Mr. Otellini has focused on driving the company’s growth and mission to deliver secure, connected, and energy-efficient computing solutions.

On October 6th, Mr. Otellini talked about innovation and Intel’s come from behind strategy to win market share in mobile computing.  The session was moderated by long time SCU EE Prof Cary Yang.

Otellini identified three challenges facing the entire technology industry:

 1.  Everywhere computing-  in phones, cars, media tablets, TVs, shopping in the mall, digital signage, etc

2.  Making the benefits of computing available to people all over the world, especially in developing countries.

 3.  Ensure innovation will continue.  Innovation was said to be “the ultimate driver of progress.”

During his prepared remarks and Q &A, Mr. Otellini made several bold statements.  Here are a few that caught my attention:

“China is more innovative than Japan.”

 “Intel is the 4th largest software company in the world, measured by number of software engineers.”

 “WiMAX was not the winning mobile wireless technology, but it drove 4G, which is LTE.” 

Analysis:  This oft cited comment really means that the pressure WiMAX proponents (like Intel) put on making OFDMA based mobile WiMAX technology available, caused the LTE camp to accelerate their own deployments.  While that is certainly true, it didn’t benefit Intel at all since the company never had an LTE development or ecosystem building program (like they had for mobile WiMAX and WiMAX 2.0).

In answer to a question posed by this author about Intel’s failure to gain market traction in mobile computing, Mr. Otellini replied that “Intel is inside many tablets, but not the iPAD.”  And that they would be announcing several “design wins” in smart phones next Spring.  He referred to a keynote talk he  will make at the January 2012 CES about Intel’s mobile computing initiatives and offered me a free ticket if I wanted to attend!

A video recording of Mr. Otellini’s speech is available at:

Deutsche Telekom (DT) Expands its Presence in Silicon Valley: 4 Focus Areas Indentified

This article was written by Louis Schreier,VP, Portfolio Development and Technology, Products & Innovation, T-Labs US.  The article was edited by Alan J Weissberger.

Having established an Industrial R&D Lab in Palo Alto in 2008, Deutsche Telekom’s (DT) T-Labs has expanded its mission in the Silicon Valley.  T-Labs already has ties to Stanford, UC- Berkeley, UC-Santa Cruz, UIUC, U Michigan and other universities.  The Labs is expanding its activities to focus on four key initiatives:

1.  Transforming its academic and industrially oriented R&D into innovative products;
2.  The integration of start-up technologies into DT’s and T-Mobile’s networks, as well as DT’s Data Centers;
3.  The incubation of promising technologies to the mutual benefit of both the startups and DT; and
4.  Creating open interfaces and APIs that enable 3rd party applications and services to take advantage of Deutsche Telekom and partner networks for consumer oriented applications as well as for B2B, B2C, and B2B2X development.

T-Labs Advanced Technology & Development Center’s internal R&D activities and charter includes Advanced Network Infrastructure, Media and Cloud-based Services, Mobile Client Ecosystem, the Development of Enabling Platforms and APIs, and Technology Integration and Incubation. Within this broad umbrella, T-Labs ATDC is pursuing several areas, including:

-Software Defined Networks (SDN);
-Android-based services and applications for mobile clients.  These include: home, health and fitness, and location-based services;
-Sensor networks and communication supporting geophysical as well as device and connected device applications;
-Media including multi-device and multi-screen video clients; and
-Cloud-based services including video conferencing and telephony services for SMBs.

Here’s a brief overview of a few of T-Labs Advanced Development Center’s activities.

1. Advanced Networking:  Although the trend for Telecommunication Operators is increasingly toward upper layer applications and services, DT and T-Labs Advanced Development Center (T-Labs ATDC) have been at the forefront in developing advanced network architectures and capabilities. As an integral part of the Clean Slate, Software Defined Networking initiative, T-Labs ATDC worked together with Stanford University.  Two T-Labs ATDC Research Scientists collaborated with and contributed to the OpenFlow specification and 1.0 reference implementation.

OpenFlow physically splits the data plane from the control plane, thereby flattening the network.  This allows flow-based monitoring and management of network traffic, which results in better network utilization of resources and more efficient control traffic distribution and dynamic migration. Commercial versions of the OpenFlow stack have been implemented by network equipment vendors, such as NEC, Juniper, and HP among others.  Nicira, and BigSwitch have implemented commercial versions of the OpenFlow controller.

The use of OpenFlow and services on-top of the OpenFlow stack in campus networks and data centers will simplify network complexity at the traffic aggregation and access network layers (L2/L3).  By splitting the control and data planes and employing less expensive hardware at the forwarding/routing layer, OpenFlow also shows promise in reducing capital and operating expenses. T-Labs ATDC is developing data center services on top of OpenFlow that support network advanced monitoring and management.

2. Mobile applications and location-based services: This example combines machine-to-machine (M2M) communications, on-board device management, and location-based services. T-Labs ATDC, together with UC Berkeley, are developing the capability to turn smart phones into a vast sensor network.  It will have the ability to detect and communicate on-board sensor states to networked servers which will provide advanced analysis to earthquake detection and alerting networks.

UC Berkeley’s focus on iOS-based devices and T-Labs ATDC focus on Android-based devices, provides the largest possible set of smart phone devices for this purpose.  The result is a mobile sensor network with the ability to contribute, in a quantifiable way, to a detection event and thereby build a picture of an earthquake’s distribution and path of direction. This is possible because the speed at which an earthquake travels is much slower than the speed of Internet communications.

If a monitoring system could shave tens of seconds to a minute off public alerts, it would provide those in the path of the earthquake and first responders key event and location-based information.  Consequently, first responders could more effectively marshal their response, and potential victims could likewise seek safety. In reality, this application is anything but simple. It involves device and communication management, on-board and server-side analytics to deduplicate, disambiguate, recognize patterns, and validate signals across potentially millions of devices communicating events in real time, in near real time.  Although this application may never be material to DT’s bottom line, its important because it serves the public good.

3.  Cloud-based business services: A final example of work T-Labs ATDC is engaged in is cloud services for SMBs. We’re currently working with two startups and the open source community to develop cloud based communication services for SMBs.  These include: conferencing, advanced PBX and telephony, as well as multiplatform videophone services.

The goal of this activity is to create advanced, and branded, services for our business customers throughout Europe, by taking advantage of innovation and open source software and services in Silicon Valley (and elsewhere). At the same time, the startups with whom we’re working are pursuing other markets and customers through their own channels and business development activities.

In this activity, T-Labs ATDC provides integration support for the core technologies as well as access to DT resources that would be difficult to obtain at so early a stage. Our goal is to provide an opportunity for DT to extend its existing consumer and business portfolio with advanced applications and services and, at the same time, provide start-ups with entry into the European Markets.

It’s important to note that when T-Labs ATDC works with startups, including those for which we provide venture funding, it’s done in a non-exclusive relationship. As a result, everyone wins!

In addition to the above development activities above our charter includes: Software Defined Networking, Middleware and Enabling platforms, OTT media, Cloud-based services for both consumers and businesses, Home technology, including home-hubs, and entertainment, Integration and Incubation.

While T-Labs Advanced Development Center is focused on building and leveraging our own resources and relationships in the valley, we’re international, and collaborate with our Innovation Labs colleagues in Germany, Israel, and our Business Unit partners in Germany and the US.

Summing up, T-Labs Advanced Development Center’s activity and collaboration with startups in the Bay Area has been ongoing for several years. Deutsche Telekom also has Venture Capital, Business Development and Strategic Partnering activities in Silicon Valley as well.

EDITORS NOTE: T-Labs will be co-sponsoring the IEEE ComSocSCV Oct 12 meeting on Cloud Networking along with Juniper Networks.   Representatives from Savvis/Century Link, Juniper, and Arista Networks will be presenting their perspective and opinons about the network that must deliver cloud compute and storage services.

Meeting information and RSVP at:


  • Michael Cowburn, Sr. Solutions Architect, Savvis
  • Omar Smith, Network Solutions Architect, Savvis
  • Anshul Sadana, VP Customer Engineering, Arista Networks
  • Colin Constable, VP/CTO Strategic Alliances, Juniper Networks

Session Organizer

Alan J Weissberger, IEEE ComsocSCV Chair and Manager of ComSoc Community web site

Session Chair

MP Divakar, IEEE ComsocSCV Secretary