Wednesday, September 24, 2014

Blueprint: The Third Network

by Nan Chen, MEF President

The Internet is everywhere, and available on-demand – but it offers no service guarantees, and leaves security to the user. CE 2.0 delivers assured performance and security levels – but it can take weeks to establish a service. The world needs a THIRD Network.

  In 2012, Ethernet bandwidth for business overtook all legacy WAN technologies, and it continues to surge ahead – Fig 1 Also, according to Infonetics Research, Carrier Ethernet revenue in services and equipment will reach $350 billion cumulatively during the next 4 years.  Effectively this means that Carrier Ethernet (CE 2.0) is now the universal “glue” holding global business together – a role once provided by the PSTN during the 20th Century.

 Just consider the two cases: the reliability, performance levels and relative security provided by today’s business services connecting offices across the globe, then compare that with the service mobile workers can expect in their own homes, hotels or on the move when using the Internet.

Of course the miracle of the Internet is that the latter services can still be immediately available almost anywhere, via fixed line or mobile connection, but the downside is that premium office services raise expectations of performance and Quality of Service (QoS) that make mobile business seem slow and frustrating by comparison. It is hard to maintain really effective business negotiations, or technical discussions, when the video image keeps freezing while the speaker talks on – let alone deal with the extra burdens of encryption and security required for safer business.

In both cases the real issue for the user is “connectivity service”. In the former case the service quality, performance and security is outstanding once the service has been established – a process that can take weeks, especially if the connection spans more than one service provider network. In the latter case the service is ubiquitous, immediate and on-demand, but falls short on quality, performance and security.

This is the challenge MEF is now addressing: to envision a THIRD Network that combines the best of both; a THIRD Network that delivers business-class connectivity as a ubiquitous, on-demand service – rather than as a laborious and lengthy installation and commissioning process.

In the words of Bob Metcalfe, Inventor of Ethernet, Professor of Innovation at the University of Texas and Advisory Director of the MEF: “Today we are embarking on the next stage of this remarkable journey with a new vision of the future of network services, building on CE2.0 capabilities. MEF is announcing a new network paradigm, a new THIRD Network vision for Agile, Assured, and Orchestrated services worldwide, one which is based on emerging Network as a Service principles”.

A fertile ground for innovation

It is all very well to express a vision for the future, but is the environment ready and able to make the vision a reality? It was 2010, four years ago, when I gave a presentation comparing Ethernet with the global PSTN, calling for: “a public service Ethernet network bound to provide Ethernet connectivity across the globe as easily as making a telephone call”.

A lot has happened since I expressed that long-term vision. For a start CE 2.0 E-Access provision has made it so much easier to extend services across multiple carrier footprints – the service is increasingly ubiquitous, although it still takes time to establish business SLAs.

Then there is market preparedness: as I write this we already have over seventy companies offering over three hundred CE 2.0 certified services and products – and that is merely since last year’s launch of the certification program – and this number is fast expanding. The importance of this technology is no secret to the business world – we already have over two thousand five hundred MEF Certified Professionals across sixty-five countries worldwide, and that represents a tenfold increase since the launch of CE 2.0.

Next there is the question of need. Humans can be very adaptable, and mobile workers have so far been managing to cope pretty well with the Internet connectivity available, considering its limitations, but the pressure for reform is mounting. It is not just the business-class expectations raised by CE 2.0, there is also the pressure of cloud adoption. As a universal solution – and that ultimately is a vital part of the Cloud’s value proposition – cloud computing will demand all the agility, assuredness and orchestration promised by the THIRD Network. Then there is M2M communications as the next “cloud on the horizon” – while human workers may adapt to whatever best effort service is available, machine to machine communications will demand automatic and immediate delivery on demand.

The THIRD Network

To summarise the argument, business today is served by two types of data networks:
·         Virtual Private Networks based on CE 2.0, MPLS etc., delivering services with assured performance and security, but taking days if not months to initiate across multiple providers.
·         The Internet, delivering on-demand, ubiquitous services but leaving users to deal with security and performance issues.

The MEF believes that there is a need for a new THIRD Network, delivering Agile, Assured and Orchestrated services worldwide, where:
  • Agile means delivery of real-time on-demand services.
  • Assured means delivery of performance and security guarantees
  • Orchestrated means automated delivery of services across multiple service providers.

 How is MEF planning to realise this vision? The solution begins with a simple question: “What makes it so hard to deliver the above type of service today?”.

The answer is that in today’s telecom operations we do not yet have “layered abstraction”. In computing, Hardware is separated from Operating Systems and from Applications, so that each layer can innovate without impacting layers above or below. In telecom operations, however, different systems are in functional silos – with separate inventory, provisioning performance and fault management systems, for example. Each such functional system must hold detailed information of all domains – like Optical, Ethernet or IP. A change in any one domain would result in changing all systems – and that takes extraordinary effort, time and money (see Fig 2)

But using Network as a Services (NaaS) principles, we can develop a layered approach for telecoms operations, to break down complex problems, domain by domain, layer by layer. To achieve this, the industry is working together in three major initiatives – NFV, SDN and Service Orchestration – to develop more responsive and adaptable solutions.

NFV allows us to use general purpose VMs instead of vendor specific network elements, for every network function. SDN is moving control functions in network elements into central controllers. Both NFV and SDN, while providing fundamental building blocks for the THIRD Network, are concerned with network element functions and control – but not the connectivity services businesses or individuals actually purchase. This is where the MEF’s Standardized Service Orchestration comes in – (Fig 3).

Service Orchestration manages the entire lifecycle of connectivity services: Fulfillment, Control, Performance, Assurance, Usage, and Analytics. It holds detailed service inventory of all services in a layer or domain, providing the necessary APIs for information exchange between service providers and internal systems operating at other layers.

Such Service Orchestration is the MEF’s immediate objective. Initial work so far includes: the definition of Service Orchestration lifecycle elements; extending information models to be dynamic and protocol independent; and developing open-standards-based APIs where needed.  All new capabilities leverage and build upon CE 2.0 as the services platform, and this ensures an optimal migration strategy.

It is important to remember that this new THIRD Network can only come into being if we work closely with many and diverse industry leaders. So an expanded collaboration program has been established, called “Unite”, involving key industry standards development bodies. Unite is critical to the acceleration of deployment and realization of the anticipated benefits for all stakeholders, for the THIRD Network – much like CE 2.0 and the Internet/IP Networks – will profoundly change the way we work, live and play.

How we will work, live and play

First imagine the impact of the THIRD Network on anyone taking a business trip. Just by connecting, whether in a hotel or on the move, your personal NaaS performance profile will be automatically loaded, a business-class connection automatically setup – and you are ready for business.
Performance-assured network experience will be available to millions of users with enormous market potential – an exciting prospect for all stakeholders.

Next consider an Enterprise network where customer orders, placed via APIs or portals, initiate automated instructions that flow through to interconnected operators – provisioning virtual service connections where available and physical installations at remote user sites where not. This shows the power of standardized service orchestration, it creates a pervasive virtual network as a platform for on-demand businesses services.

Cloud service delivery will become seamless, with service on-demand network services that interconnect user locations to their virtual machines or Virtual Network Functions running on blade servers inside a remote data center. Automated inter-carrier service orchestration will manage all connectivity – inside and outside the data center.

Finally, to give a taste of what the THIRD Network could mean for society beyond the world of business, I refer back to that presentation I gave in 2010, where I also introduced a concept I called ”Personal Ethernet”.

I gave a single household as an example: Dad, is a media executive needing massive bandwidth to download hi-def video samples; Mum is a social worker, highly reliant on VoIP for counselling and departmental conference calls; Daughter needs reliable fast Internet for schoolwork; Son is crazy about on-line gaming and needs massive bandwidth plus minimal latency and packet loss.

The THIRD Network could deliver all these personalized requirements “as a Service” that would be available equally at home, at work, in school… or even while on vacation. Personal Ethernet would be the ultimate prize for fixed and mobile convergence, and the THIRD Network will make it a practical proposition.

As Bob Metcalfe said in summary at the launch of this vision: “Once again we stand at a new beginning for the world of networking. The THIRD Network auto-orchestrates dynamic, performance-assured services and brings significant growth in the market for all stakeholders. Wouldn’t it be great to have Network as a Service personalized, virtualized and mobilized with an SLA that meets your every need? That’s the promise of the new THIRD Network.”

About the Author

Nan Chen

Nan Chen

President, MEF
Executive Vice Chairman, CENX
Nan is known in the telecom/networking industry as the founding President of the MEF. The MEF has profoundly shaped the telecom and Internet industry by establishing Carrier Ethernet as the predominant technology and service, reaching $60b in revenue, for businesses, homes, mobile infrastructure, and for next generation networking of the Internet.

About the MEF

A communications industry alliance consisting of more than 225 member organizations, the MEF is the defining body for Carrier Ethernet and the driving force behind the global market for Carrier Ethernet services & technologies.

The MEF operates through a powerful collaborative framework where service providers, network solutions suppliers, and other member companies contribute to the development of technical and operational specification, certification, and marketing programs. The MEF has enabled Carrier Ethernet adoption and tremendous communications-related cost savings and productivity gains for service providers and end-users worldwide over the past decade.

Building on this success, the MEF is now focused on enabling a new generation of agile, assured connectivity services that are orchestrated over more efficient, automated, and interconnected global networks. For more information about the MEF – including details about services, network equipment, and professional certification – please visit

Broadcom's StrataXGS Tomahawk Switch Aims for Cloud Data Centers

Broadcom introduced its next-generation, StrataXGS "Tomahawk" switch, packing 2 Tbps of switching performance in a single chip that is capable of supporting 32 ports of 100 Gigabit Ethernet, 64 ports of 40GE/50GE or 128 ports of 25GE.

The new chip, which represents the 7th generation of Broadcom Ethernet switching silicon, is designed for the programmable fabrics needed in next-gen, cloud-scale data centers.

The StrataXGS Tomahawk Series is powered by more than 7 billion integrated transistors in 28nm. It supports new 25GE and 50GE protocol standards, and is the first chip to scale to 32-ports of 100G on a single device.

Significantly, Tomahawk implements Broadcom's FLEXGS and Smart-Flow technology, enabling the switch to be software configured for specific policies while preserving deterministic latency. The new design brings a 12X increase in application policy scale compared to previous generation switches, along with increased flexibility of packet lookups and key generation, and rich load balancing and traffic redirection controls. This supports an extensive suite of user configurable functions for flow processing, security, network virtualization, measurement/monitoring, congestion management and traffic engineering.

"Our StrataXGS Tomahawk Series will usher in the next wave of data centers running 25G and 100G Ethernet, while delivering the network visibility required to operate large-scale cloud computing, storage and HPC fabrics," said Rajiv Ramaswami, Broadcom Executive Vice President, Infrastructure & Networking Group. "This is the culmination of a multi-year cooperative effort with our partners and customers to prepare for this transition. We are pleased to see significant industry investment in the Tomahawk 32x100GE form factor as well as the 25G/50G Ethernet specification, which Broadcom defined and co-founded as an industry standard."

StrataXGS Tomahawk Key Features
  • 3.2 Tbps multilayer Ethernet switching
  • Integrated low-power 25Ghz SERDES
  • Authoritative support for 25G and 50G Ethernet Consortium specification
  • Configurable pipeline latency enabling sub 400ns port-to-port operation
  • Supports high performance storage/RDMA protocols including RoCE and RoCEv2
  • BroadView instrumentation: provides switch- and network-level telemetry
  • High-density FleXGS flow processing for configurable forwarding/match/action capabilities
  • OpenFlow 1.3+ support using Broadcom OF-DPA
  • Comprehensive overlay  and tunneling support including VXLAN, NVGRE, MPLS, SPB
  • Flexible policy enforcement for existing and new virtualization protocols
  • Enhanced Smart-Hash™ load balancing modes for leaf-spine congestion avoidance
  • Integrated Smart-Buffer™ technology with 5X greater performance versus static buffering
  • Single-chip and multi-chip HiGig™ solutions for top-of-rack and scalable chassis applications
The device is currently sampling to lead customers and cloud providers.

In this 9-minute video interview, Nick Ilyadis, VP/CTO of the Networking Products Group at Broadcom, discusses the evolution of data center switching.

Key topics covered include:

00:08 - How are data center traffic patterns changing due to Big Data apps and scale-out clouds?

01:05 - How will data center networking evolve to accommodate cloud services across the WAN?

02:48 - What is the case for 25G/50G Ethernet?

04:10 - How do you see the NFV ecosystem emerging?

07:08 - What does the current trajectory of cloud services and hyper-scale data centers tell you about the future of switching?

08:28 - What's next for Broadcom?

In April 2014, Broadcom expanded its XLP II multi-core communications processors family designed for Network Functions Virtualization (NFV) and Software Defined Networking (SDN) equipment. The XLP500 processor features 32 NXCPUs with an innovative quad-issue, quad-threaded superscalar architecture with out-of-order execution for 80 Gbps performance. The processor supports Broadcom's Open NFV platform provides seamless interoperability with Broadcom's StrataXGS Switch Series.

XLP500 Series Key Features:
  • Optimized driver and tight integration with Broadcom's FASTPATH software enables seamless interoperability with Broadcom's StrataXGS Switch Series
  • Support for Broadcom's Open NFV platform enables seamless migration of virtual functions independent of Instruction Set Architectures (ISAs)
  • Full end-to-end virtualization of CPUs memory, I/Os and acceleration engines
  • Integrated autonomous acceleration engines support a broad range of applications such as Deep Packet Inspection (DPI), encryption, network acceleration and storage
  • Configurable pool of high performance multi-core processors along with application-specific engines accelerates advanced networking and security functions in a virtualized environment
  • Supports open-source virtualization platforms such as KVM, QEMU and OVS
  • Integrated high speed I/Os including 40GEs, 10GEs, GE, HiGig2 and Interlaken
  • Dedicated lanes for PCIE Gen3.0 and SATA3.0

In August 2012, Broadcom introduced its new StrataXGS Trident II Series Ethernet switching chip designed for extremely dense, cloud-scale data centers.  The design brought 2X greater forwarding density, a 4X increase in network virtualization scale, and the ability to deliver 100 10GbE ports for massive scale in top-of-rack or aggregation switches. Uplink capabilities in the flexibility to support up to 32 40GbE ports.

With this release, the company also introduced SmartSwitch" technologies, including

  • Smart-NV: Delivers up to 4X greater scale, enabling SDN virtualization across both virtual and physical workloads at wire-speed. Enables cloud-scale network infrastructure virtualization using advanced Layer 2 over Layer 3 (L2oL3) network virtualization technologies such as NVGRE and VXLAN.
  • Smart-Buffer: Delivers up to 5X higher packet buffer utilization and burst absorption performance using innovative traffic load-based intelligent and dynamic allocation schemes.
  • Smart-Table: Delivers highest Layer 2 and Layer 3 forwarding scale with network topology-based profiling for maximum deployment flexibility.
  • Smart-Hash: Eliminates polarization and load imbalances in fat-tree networks possessing heavy and diverse traffic patterns. Provides unprecedented network traffic visibility and diagnostics required by software defined networking. 
Sampling of the StrataXGS Trident II Series is expected this quarter with multiple device options.

Broadcom Interview: What's Next for Data Center Switching

In this 9-minute video interview, Nick Ilyadis, VP/CTO of the Networking Products Group at Broadcom, discusses the evolution of data center switching.

Key topics covered include:

00:08 - How are data center traffic patterns changing due to Big Data apps and scale-out clouds?

01:05 - How will data center networking evolve to accommodate cloud services across the WAN?

02:48 - What is the case for 25G/50G Ethernet?

04:10 - How do you see the NFV ecosystem emerging?

07:08 - What does the current trajectory of cloud services and hyper-scale data centers tell you about the future of switching?

08:28 - What's next for Broadcom?

Alcatel-Lucent Ships 100G/200G Single Carrier Line Card

Alcatel-Lucent announced commercial delivery of a single-carrier 100G/200G DWDM optical line card for its 1830 Photonic Service Switch (PSS) platform. This allows operators to transmit data over distances as great as 4,000 km in 100G mode and as far as 1,000 km in 200G mode.

The 200G solution initially allows for the doubling of fiber capacity from 8.8 Tbps (88 x 100G) to 17.6 Tbps (88 x 200G). Alcatel-Lucent’s flexible grid feature delivers a total system capacity of 24 Tbps.  The solution is backward compatible with previous generations of Alcatel-Lucent’s 100G DWDM solution.

The 100G/200G single carrier line card has already been deployed into 20 service providers networks.

Accuris Raises $15 Million for Roaming and Interworking Connectivity

Accuris Networks, a start-up based in Dublin, Ireland with offices in the U.S., Malaysia, Brazil and Japan, completed a $15 million equity funding round for its platform for connecting current and emerging mobile broadband networks such as GSM, CDMA, LTE and Wi-Fi.  The company also announced the appointment of Jeff Brown as CEO.

The Accuris AccuROAM platform, which includes smartphone client software, access server/gateway technology and billing integration support, helps service providers interconnect various network technologies and domains for a better subscriber roaming experience.  The platform can be used for Wi-Fi Offload and Wi-Fi Roaming, including Wi-Fi Calling and Wi-Fi Texting. This enables carriers to include Wi-Fi as a full service-capability alongside 3G/4G Mobile, fixed broadband, etc.

The newly named CEO Jeff Brown previously was president and CEO of mobile technology company Kineto Wireless. He held similar positions at RadioFrame Networks and was earlier the CEO of Data Critical, a wireless healthcare company.

Key new investors in this privately held company are the Ulster Bank Diageo Venture Fund and the China Ireland Growth Technology Fund.

"Our established industry customer base, our significant pipeline and our industry-leading technology were all factors that attracted this investment. We intend to use this funding to invest in future product development in our AccuROAM™ platform, to maintain our lead in connecting networks and build more solutions such as carrier-grade Wi-Fi offload and roaming as we scale our business globally. While we already have customers worldwide and support from existing investors such as Atlantic Bridge Capital, we particularly want to expand in China, Japan and other Asian markets. We also intend to consolidate and amplify our successes in EMEA, particularly in the Middle East and Western Europe," stated Accuris Networks Chairman Larry Quinn.

  • Accuris Networks was founded in 2003 by Aidan Dillon, who is now the company's CTO. 

SiTime Raises $25 Million for MEMS Timing

SiTime Corporation, a MEMS analog semiconductor company based in Sunnyvale, California, has closed $25 million in new financing, which consisted of a combination of structured debt facility of $15 million provided by Capital IP Investment Partners and strategic equity investment from other investors.

SiTime specializes in silicon MEMS-based oscillators and clock generators that are a drop-in replacement for legacy quartz crystal products. The company has shipped over 200 million devices to date.

“SiTime has had a strong and enduring vision – to revolutionize the timing market with MEMS and analog technologies. Today, that vision is a reality, achieved through single-minded focus and rapid innovation. We have the best MEMS technology, the broadest product portfolio, the largest customer base, the highest shipped volumes, and are the preferred timing supplier for exciting new applications such as wearables and Internet of Things (IoT),” said Rajesh Vashist, CEO of SiTime “This financing, along with our leadership position, forms an unbeatable combination for future growth and innovation. We are pleased that CapIP recognizes SiTime’s market and revenue leadership and we value their industry expertise and ability to work with us to tailor the financing to our business objectives and long-term strategy.”

In 2013,  SiTime introduced a family of 32 kHz MEMS oscillators designed to replace legacy quartz crystal resonators in mobile devices, such as smartphones, tablets, fitness monitoring and watch products, medical monitoring and wellness devices, etc.  

As with the company's other MEMS timing solutions, the new SiT15xx family of oscillators compare favorably to existing quartz-based devices in several ways; SiTime’s MEMS are 85% smaller, use 50% less power, and significantly exceed the reliability metrics of quartz by 15 times.

Alcatel-Lucent Kicks Off Sea-Me-We 5 Construction

Alcatel-Lucent confirmed that its construction with the South East Asia-Middle East-Western Europe 5 (SEA-ME-WE 5) consortium has come into effect.

The new 100G SEA-ME-WE 5, which spans 20,000 km from Singapore to France, boasts an ultimate design capacity of 24 Tbos.

Ms Linette Lee , Chairperson of the SEA-ME-WE 5 Management Committee said:  “With the rollout of the SEA-ME-WE 5 system,  we will deliver a dynamic and agile way to serve end-users and offer them the quality of service they need. Alcatel-Lucent and NEC’s 100G undersea technology will help us in setting a new milestone in the further development of this route.”

Alcatel-Lucent said it will supply its 1620 Light Manager submarine line terminal using Advanced Coherent technology and incorporating the latest Soft Decision Forward Error Correction (SDFEC) technology, as well as its advanced ROADM branching unit offering dynamic features for enhanced system resilience. Alcatel-Lucent will also install cable and repeaters, and the cable system will be managed by Alcatel-Lucent’s end-to-end submarine network management system.

  • In March 2014, Alcatel-Lucent and NEC both announced contract wins for the new SEA-ME-WE5 project, a 100G undersea cable system that will span 20,000km from Singapore to Europe.
    The South East Asia-Middle East-Western Europe 5 (SEA-ME-WE 5) project is backed by consortium of 15 leading telecom operators.
    Alcatel-Lucent will be deploying the segments spanning from Sri Lanka to France. Co-contractor NEC will be deploying the Singapore to Sri Lanka segment.

Gigamon: Unified Visibility Needed for VoLTE, 100G, Network Virtualization

Gigamon, which offers active visibility solutions based on its Unified Visibility Fabric, announced a strategy to help network service providers de-risk the rollout of three key technologies: VoLTE, 100G transport links and carrier network virtualization.

Gigamon said it is uniquely positioned to address this “Service Provider Triple Challenge of Network Transformation”  because its Visibility Fabric provides operators with the capabilities to understand both the traffic and technology interdependencies.

“Gigamon is seeing carriers begin deployments with one of the technologies and end up deploying all three in parallel, which causes an unexpected resource crunch that only Gigamon's Visibility Fabric can resolve through the deployment of a unified tool rail. This quickly becomes apparent with two of the new technologies, 100Gb and network virtualization, since they actively remove monitoring visibility.” said Andy Huckridge, director of service provider solutions at Gigamon.

“By deploying a unified tool rail and driving it with the correct traffic, carriers can better understand the interdependencies when deploying these new technologies. This allows for the de-risking of all three new technologies, regardless of which one is deployed first – and keeps resource needs in check while allowing more than one technology to be deployed at the same time. The Visibility Fabric can help carriers understand how each of the Triple Challenge technology deployments affects each other, how each transported traffic type is affected by each deployed technology and how Real Time Protocol (RTP) traffic is affected by other transported traffic types on the same pipe.”

HP Sets Carbon Reduction Goals

HP announced a goal to reduce the emissions intensity of its product portfolio by 40 percent by 2020 compared to 2010 levels.  The carbon reduction goals apply to all three parts of its value chain—operations, supply chain and product portfolio.

“As one of the world’s largest IT companies, we believe we are uniquely positioned to help our company and our customers lower carbon emissions by developing more sustainable technologies that replace outdated, inefficient processes and behaviors,” said Gabi Zedlmayer, vice president and chief progress officer, Corporate Affairs, HP. “With this new emissions goal, we are once again demonstrating our commitment to building a low-carbon economy that creates a better future and a healthier world for generations to come.”

The company cited its work in improving data center efficiencies by developing energy-efficient solutions such as the HP Moonshot server architecture, which consumes up to 89 percent less energy, uses 80 percent less space and costs 77 percent less than a traditional server environment.