Wednesday, May 2, 2018

Open Disaggregated Transport Network project gets underway

A new, operator-led Open Disaggregated Transport Network (ODTN) project is underway at the Open Networking Foundation (ONF).

The goal of ODTN is to build optical transport networks using disaggregated optical equipment, open and common standards, and open source software.

The project will deliver an open source platform for running multi-vendor optical transport networks. It will leverage the ONF’s ONOS SDN Controller, automatically and transparently discovers the disaggregated components and will control the entire transport network as a unified whole, thus enabling multi-vendor choice.

The organizers of the project say that just as the SDN movement has disaggregated the data center and operator edge networks, ODTN will bring similar benefits to the optical transport network including best-of-breed choice, elimination of vendor lock-in, cost containment and accelerated innovation.

Backers of the project include China Unicom, Comcast, NTT Communications, Telefonica and TIM.

Each of the five founding operators has committed to performing lab integration and evaluation of the platform for future transport applications. Additional support is coming from leading vendors in the optical equipment space, with NEC, NOKIA, Oplink, ZTE contributing to the software platform and building full solutions, CTTC contributing from academia, and ADVA, Ciena, Coriant, CoAdna, Infinera and Lumentum participating in lab and field trials.

Relationship to Other Projects

ODTN is the only open source solution in the optical transport space, but is leveraging other ongoing work which has focused on standardizing various interfaces and components.

ODTN will leverage and expose TAPI as its northbound interface, leveraging the work coming out of the ONF’s Open Transport Configuration and Control (OTCC) project. Likewise, OpenConfig is the base southbound model and API for communicating to optical equipment.

The OpenROADM MSA defines interoperability specifications and data models for optical devices, networks and services.  ODTN benefits from this effort and, over time, it helps the industry achieve transponder compatibility.  This will eliminate the need to deploy transponders in matched pairs, further disaggregating the solution and enabling even greater deployment flexibility.

TIP’s Open Optical & Packet Transport project is producing open DWDM architectures, models and APIs, covering transponders, open line systems, and routers. In time, the ODTN project hopes to benefit from the availability of open optical hardware coming from the TIP work.  And visa versa, the TIP project can leverage the open source work coming out of ODTN on TIP white box hardware building blocks (such as Voyager).

“It is one of the most innovative technical challenges to deploy open SDN / Disaggregation technologies into transport networks. We expect that it will dramatically shorten the service development term and reduce costs,” said Dai Kashiwa, Director of NTT Communications and an ONF board member representative of the NTT Group. “The reference design and implementation for ODTN will accelerate this challenge, and provide common usefulness among many service providers. So, we are so excited that many service providers and vendors have aligned with the ODTN concept, and started collaboration on specifying common requirements and test/deployment plans. We aim to build and nurture an ecosystem that allows us to deploy and operate ODTN-based production networks.”

Coriant announces CEO change

Coriant named Pat DiPietro as its new Chief Executive Officer, effective immediately, replacing CEO and Chairman Shaygan Kheradpir, who has stepped down from his role to pursue other opportunities.

DiPietro will continue to serve as Vice Chairman of the Board, a role he has held since the founding of Coriant in 2013. DiPietro previously held senior leadership roles at Nortel and Bell Northern Research. He also previously served as Managing Partner at Canada’s VG Partners, overseeing the company’s Technology Fund. As a venture capitalist, he managed large portfolios and teams and sat on numerous Boards, including Sandvine, SiGe, Continuous Computing, BTI Systems and BelAir Networks.

Coriant also announced that Reza Ghaffari has been promoted to the role of Chief Operating Officer (COO), a new position within Coriant. Ghaffari will continue to lead Coriant’s global service and support organization, while assuming responsibility for the company’s global IT, human resources, and facilities functions. Between 2000 and 2005, Ghaffari also worked at Verizon where he was responsible for innovation, product development, and strategic partnership programs. Dur

“On behalf of the Board and the management team, we wish to thank Shaygan for his commitment and many contributions to Coriant over the past three years,” said DiPietro. “I’m thrilled to take the helm of Coriant as it transforms to drive new value for its global customers with cost-disruptive innovations in open, software-driven, and revenue-enhancing products and technologies.”

CyrusOne hits year-over-year revenue growth of 32%

CyrusOne, a data center REIT, posted Q1 2018 revenue of $196.6 million, up 32% over the same period last year. The increase in revenue was driven primarily by a 29% increase in occupied CSF, lease termination fees totaling $5.0 million, and additional interconnection services.

The company said it leased approximately 29 MW of power and 226,000 CSF in the first quarter, representing $3.4 million in monthly recurring rent, inclusive of the monthly impact of installation charges, or approximately $40.4 million in annualized GAAP revenue5, excluding estimates for pass-through power.

In the Northern Virginia data center market, CyrusOne is leasing space as fast as it can add it.


Lumentum posts revenue of $298.8 million

Lumentum reported net revenue for the fiscal third quarter of 2018 of $298.8 million, with GAAP net income of $2.7 million, or $0.04 per diluted share.

Non-GAAP net income for the fiscal third quarter of 2018 was $50.6 million, or $0.78 per diluted share. Non-GAAP net income for fiscal second quarter of 2018 was $107.8 million, or $1.67 per diluted share. Non-GAAP net income for the fiscal third quarter of 2017 was $30.8 million, or $0.49 per diluted share.


"Our strategy of investing in differentiated products and technologies, focusing on close relationships with market leading customers, and leveraging our technologies across multiple growing end markets, is working.  Driven by strong customer demand and execution on capacity expansion, in the third quarter we achieved new record Lasers revenues, which increased 18% sequentially, and grew Telecom revenues by more than 11% sequentially, with notable strength in ROADMs, which were up 27% sequentially," said Alan Lowe, President and CEO. "Though seasonally down, we made good progress on new 3D sensing customer programs and are well positioned for new customer product introductions during FY19. During the third quarter, we announced reaching an agreement to acquire Oclaro and we continue to work with Oclaro on this pending transaction."


Lumentum to acquire Oclaro for $1.8 billion

Lumentum agreed to acquire Oclaro for approximately $1.8 billion in cash and stock.

Under the deal, Oclaro stockholders will be entitled to receive $5.60 in cash and 0.0636 of a share of Lumentum common stock for each share of Oclaro stock, representing a premium of 27% to Oclaro's closing price on March 9, 2018 and a premium of 40% to Oclaro's 30 day average closing price.  Oclaro stockholders are expected to own approximately 16% of the combined company at closing.

The combined company is expected to have annual revenue of $1.733 billion and an operating margin of 19%, prior to synergies from the combination.

Lumentum, which is based in Milpitas, California, supplies a range of optical components and subsystems for telecom, enterprise, and data center networking equipment. The company was created in 2015 as a split off from JDSU.

Oclaro supplies optical components and modules for the long-haul, metro and data center markets. The company is based in San Jose, California.

Equinix revenues up 28% yoy

Equinix's quarterly revenues increased 28% year-over-year to $1.216 billion; a 10% year-over-year increase on a normalized and constant currency basis.

Peter Van Camp, Executive Chairman and Interim CEO and President, Equinix, stated: "As Equinix approaches its 20th anniversary, we are excited to post our 61st quarter of consecutive revenue growth, which is reflective of the critical role we serve in helping businesses interconnect their IT infrastructure to succeed in the digital economy. Equinix currently serves nearly half of the Fortune 500 and our recent acquisitions, combined with our currently announced organic expansions, have positioned Equinix to capture an even greater share of the market opportunity."

Orange Business Services names CEO

Orange Business Services named Dr. Helmut Reisinger as its new chief executive officer, replacing Thierry Bonhomme, who becomes special advisor to the Chairman and CEO of Orange before retirement later this year. Helmut will report to Stéphane Richard, Chairman and CEO of Orange, and will be a member of the Group’s Executive Committee.

Most recently Helmut was executive vice president, International at Orange Business Services, in charge of all international enterprise business activities, excluding France. Before joining Orange Business Services in 2007, Helmut held management positions across Europe at Avaya Inc, NextiraOne Germany and Alcatel Austria.

"I am honored and excited to lead Orange Business Services on its ambition to be at the forefront of the data-driven economy. I believe our global talent, expertise and assets position us to deliver an unmatched experience for our enterprise customers worldwide. With a relentless customer focus  – combined with people empowerment and commitment to innovation – I am confident that we will achieve continued success and growth for both our customers and Orange,” said Dr. Helmut Reisinger, chief executive officer, Orange Business Services.




Start-up profile: TidalScale, building an inverse hypervisor for scale-up servers

TidalScale, a start-up based in Campbell, California, is on a mission to build the world's largest virtual servers based on Intel x86 commodity hardware.

The company's "inverse" hypervisor combines multiple physical servers (including their associated CPUs, memory storage and network) into one or more large software-defined virtual servers. This is the inverse equivalent of VMware because a rack of physical servers are virtualized as though it were one. The concept is to scale-up a virtual server instance to handle Big Data workloads without making changes to applications or operating systems.

Why use another hypervisor to create a bigger server? Doesn’t Moore’s Law already deliver more powerful processors over time? And why not just provision a large number of individual servers from a Cloud IaaS vendor? The answers here would be (1) very large in-memory datasets (2) Moore’s law is not keeping pace with rising workloads demands (3) too costly and too limiting, especially since public cloud operators tend to limit the memory size of bare metal servers to 2TB and because in load balancing a workload there is a tendency to provision to more resources than necessary.

The TidalScale story

TidalScale was founded in 2012 by Dr. Ike Nassi, an Adjunct Professor of Computer Science at UC Santa Cruz, who has been involved in many tech developments including as Chief Scientist at SAP when the category of in-memory databases was established. He also was involved in 3 previous start-ups: Encore Computer, a pioneer in symmetric multiprocessors; InfoGear Technology, which developed Internet appliances and services; and Firetide, a wireless mesh networking company.

The technical team also includes Dr. David Reed as Chief Scientist, who holds many patents along with four degrees from MIT in EE and CS including his PhD. Reed's contributions to the networking field include work on the original Internet protocol design team. His architectural contributions included the UDP protocol design, the “slash” in TCP/IP, and formulation of the End-to-End Argument as its primary protocol design principle. Later, he went on to become  Chief Scientist at Lotus Development Corporation, an HO Fellow, and an SVP at SAO Reseach.

On the management side, TidalScale is headed by Gary Smerdon, who previously was the EVP & Chief Strategy Officer of Fusion-io, the devel.oper of flash-based PCIe hardware and software solutions that was ultimately acquired by SanDisk in 2014 for $1.3 billion. Before that, Smerdon was SVP and GM of the Accelerated Solutions Division at LSI, an internal startup that he founded. Smerdon also held executive positions at Greenfield Networks (acquired by Cisco), Tarari (acquired by LSI), Marvell, and AMD.

TidalScale, which first began shipping in 2016, aggregates all the resources (memory, cores, storage and bandwidth) of low-cost, high-performance, 2-socket Intel x86 servers into one or more Software Defined Servers. This accomplished by running a TidalScale HyperKernel on the physical server and a "WaveRunner" control plane and management console to orchestrate the spinning up or spinning down of virtualized servers. The HyperKernal instance on each physical server communicates with other HyperKernal over the Ethernet network, which essentially functions as a combined memory and I/O bus. Thus memory performance will be determined by the latency and throughput of the Ethernet connection. Still, for applications such as very large in-memory databases, a TidalScale software-defined server consisting of five physical nodes each with 128GB of DRAM, will be better than a single server with 128GB of DRAM if the memory required exceeds 128GB and a secondary SSD must also be employed. This is because DRAM performance is roughly 1000X that of flash memory.

Software-Defined Servers can be configured with dozens or even hundreds of processor cores, tens of terabytes of memory, and as much storage and networking I/O as needed. The configuration of servers can be automatically right-sized to the workload. TidalScale allows Docker containers and container management platforms (Kubernetes) to run on top. For instance, TidalScale could be used to deploy a single Linux instance with 15TB of DRAM and up to 400 cores by leveraging dozens of servers in a cloud data centre.

As mentioned above, TidalScale's paradigm scale-up paradigm on commodity servers should be especially relevant to in-memory databases, such as SAP HANA. The company says it can configure up to 64TB of in-memory performance on 2-socket Intel x86 servers. Currently, cloud customers can TidalScale to on standard servers available on IBM BlueMix, OrionVM’s Wholesale Cloud Platform, and Oracle Cloud Infrastructure, with virtual systems ranging from dozens to hundreds of cores and featuring up to 30TB or more of memory. Natural allies then would include any company in that database ecosystem. Because TidalScale was exhibiting at the Open Compute Project Summit, it reasonable to assume that it sees the hyperscale cloud companies also as potential customers.

TidalScale has received a number of awards, including being named a Gartner Cool Vendor, an IDC Innovator for 2017, a Red Herring Top 100 North America recipient for 2017. Another milestone occured in November 2016 when Infosys made an equity investment in TidalScale. Financial terms were not disclosed. Crunchbase says TidalScale has gone through several rounds of venture funding, raising at least $11.8 million, probably more.

In the broader context of software-defined data centres, the need for scale-up servers will certainly be just as important as scale-up storage. Many start-ups have pursued the JBOF (just a bunch of flash) storage array opportunity, and some of these companies were acquired at nice premiums and other completed IPOs. The software-defined server space likely won't have as many start-up entrants, giving this company a better chance at driving its inverse hypervisor paradigm forward.

Nokia sells Digital Health business

Nokia will sell its Digital Health business to Éric Carreel, co-founder and former chairman of Withings. Financial terms were not disclosed.

Digital Health's business portfolio includes consumer and enterprise products, and it manufactures and sells an ecosystem of hybrid smart watches, scales and digital health devices to consumers and enterprise partners.

Tuesday, May 1, 2018

OPNFV's Fraser platform release brings cloud-native capabilities

The OPNFV Project announced its sixth OPNFV platform release: OPNFV Fraser, advancing the state of NFV around cloud native applications and new upstream project integration while continuing end user support as they deploy and test virtualized networks.

OPNFV provides the platform and tooling required by developers to validate, integrate, onboard, and test NFVI, VIM, VNFs, and network services.

Key updates in OPNFV Fraser include:

  • Advancing the support for cloud-native NFV. Fraser expanded cloud native NFV capabilities in nine different projects, more than doubled the number of supported Kubernetes-based scenarios, deployed two containerized VNFs, and integrated additional cloud native technologies from the Cloud Native Computing Foundation (CNCF) relating to service mesh (Istio/Envoy), logging (fluentd), tracing (opentracing with Jaeger), monitoring (Prometheus), and package management (gRPC). These updates move the cloud native capabilities from basic container orchestration to include operational needs for cloud native applications. Additionally, the FastDataStacks project takes advantage of FD.io work to incorporate the VPP dataplane into Kubernetes networking capabilities to enable cloud native network-centric services.
  • More mature testing. OPNFV continues to focus on the real-world deployment needs of service providers by expanding test case coverage and scope. Testing projects in Fraser see a robust increase in test cases. Functest, the OPNFV functional testing project, now permits use of its framework with other open source projects such as ONAP. This avoids duplication, reduces VM size, and accelerates the creation of additional test cases. Functest also added test cases to cover Kubernetes and Clover and made it easier and faster to run functional tests. Also in support of real-world needs, performance test projects extended the Day 0 performance testing to long-running performance testing as Day N operational issues become more real for service providers.
  • Continuous Integration (CI) updates enable increased community hardware utilization, which in turns speeds up the testing process. Fraser includes the latest versions of upstream projects and advanced dynamic CI with the introduction of metadata descriptor specifications for Scenarios, PODs, and installers that will make hardware allocation for scenarios dynamic and automated. The XCI cross-community project made additional cloud-native strides by initiating CI/CD integration work with the CNCF Cloud CI project.
  • New carrier-grade features are added, specifically in the areas of monitoring, service assurance, networking, and dataplane acceleration. Specific new features include:
  • The Doctor project, in conjunction with OpenStack, whose collaboration was instrumental in achieving this milestone, introduced an infrastructure maintenance use case for zero VNF downtime. Similarly, Barometer continued to expand the monitored items list and plugin support. The Calipso project added support for Kubernetes and physical/physical-virtual switch connections across heterogeneous environments.
  • The SFC, SDNVPN, FastDataStack, and Parser projects added new features around networking and dataplane acceleration.
  • The IPv6 project now supports clustering, simplifying network configuration, and is exploring IPv6 container networking.


“Since inception, OPNFV has been the place for industry collaboration with upstream communities, which has grown even more with the Fraser release,” said Heather Kirksey, VP, Community and Ecosystem Development, The Linux Foundation. “With more mature cloud native integration and expanded testing and collaboration, OPNFV delivers the tools needed for end users to validate and test new network services.”

Cisco divests its Service Provider Video Software business

Cisco agreed to sell its Service Provider Video Software Solutions (SPVSS) business to a company backed by the Permira Funds. Financial terms were not disclosed.

Permira Funds will create a new, rebranded company focused on developing and delivering video solutions for the Pay-TV industry. The new company's portfolio includes Cisco's Infinite Video Platform, cloud digital video recording, video processing, video security, video middleware, and services groups. Dr. Abe Peled, former Chairman and CEO of NDS and adviser to the Permira Funds, will serve as Chairman of the new company.

Cisco will retain the video and media technology related to its core business in networking, multi-cloud, security, data, and collaboration.

"This is a unique opportunity to lead and shape the video industry during its transition with the flexibility as a private company," said Dr. Peled. "The new company will have the scale, technology innovation, and world-class team to deliver outstanding go-to-market execution, customer engagement, and new end-user experiences.  Cisco has built a profitable business in the video space with innovations to capitalize on IP distribution and cloud-based services. These combined assets provide a significant new opportunity for the new company."

Cisco's 'Videoscape Unity' TV Platform -- Streaming Under the Cloud

Cisco introduced its "Videoscape Unity" TV Platform featuring a multiscreen cloud digital video recorder (DVR), which enables consumers to restart shows, catch up on past programs, and play back DVR-captured content from anywhere, on any screen.

Videoscape Unity, which is designed for service providers and media companies, is an open software platform that was created by combining the assets of NDS (which Cisco acquired last year ) with its own Videoscape portfolio.  The new platform comprises a set of cloud, network and client based components, connected by open interfaces.  Some pre-integrated components include:
  • Multiscreen Cloud DVR: Offers cloud-driven video recording with capture and storage in the cloud instead of the end device. Consumers can restart shows, catch up on past programs, and play back DVR-captured content from anywhere, on any screen.
  • Video Everywhere: Broadens the TV Everywhere proposition with unified search, discovery, and viewing functions to allow consumers to watch premium live and on-demand content on any (service provider managed or unmanaged) connected device regardless of location.
  • Connected Video to Any Device in the Home: Cisco's Connected Video Gateway serves as a single entertainment hub, with back-end management of IP and QAM video, for distributing video content and metadata to any IP-connected device in the home, while providing a unified user experience. 
  • IP Video over Cable: Gives consumers expanded choice of content and IP video services, with faster delivery of on-demand and interactive offerings, across a wider range of service provider managed devices - with the flexibility to add unmanaged devices.

Cisco said a key advantage of Videoscape Unity is that the cloud can now be used "to power personalized video services and enable multiple screens to be synchronized to create a single unified experience for the subscriber, so things look and feel the same no matter what device they use."

Significantly, Cisco is offering Videoscape "as a service," allowing operators to have Cisco build, monitor, operate and even host their video infrastructures.

http://www.cisco.com/en/US/netsol/ns1043/networking_solutions_market_segment_solution.html


  • In March 2012, Cisco agreed to acquire NDS Group in a deal valued at approximately $5 billion. NDS, which was owned by News Corp.(49%) and Permira private equity (51%), developed video software and content security for media companies, cable & satellite TV operators and IPTV service providers. Key products included its MediaHighway Set-top Box middleware software, its "XTV" Digital Video Recorder software, its "Snowflake" electronic program guide (EPG), and its "VideoGuard CA" and "VideoGuard Connect" digital rights management system.  NDS customers include some of the largest cable, satellite and broadband pay-TV operators, including Astro, Bharti, BSkyB, Canal Plus, China Central Television ("CCTV"), Cox, DIRECTV, Kabel Deutschland, Sky Deutschland, Sky Italia, TataSky, UPC and Vodafone. The company notes that a significant portion of its business is recurring, with long-term contracts, typically with an average duration of approximately five years. NDS, which is based in the U.K., has approximately 5,000 employees with facilities in Israel, France, India and China.

  • In January 2011, Cisco's John Chambers outlined a new "Videoscape" portfolio of five major product families aimed at "transforming the TV experience." From the outset, the goal was to work with Service Providers to allow any device over any network to access any content to which they are entitled. Cisco Videoscape would enable service providers to monetize activities outside their own network or traditional device footprints.  A key facet of Videoscape is about delivering a consistent interface across multiple devices, while providing a universal guide and search capabilities across all content sources. This requires building capabilities into the service provider's video back-office using APIs extending across content management systems and virtualized storage. The capabilities would be social network-aware and open to advertising opportunities.

Cisco to acquire Accompany for $270 million

Cisco, agreed to acquire Accompany, a start-up developing an AI-driven relationship intelligence platform, for $270 million in cash.

Accompany, which is based in Los Altos, California, offers business insights for finding new prospects, navigating the selling process, and strengthening relationships. Accompany Founder and CEO Amy Chang will join Cisco as senior vice president in charge of the Collaboration Technology Group. Chang, who has served as a member of Cisco's Board of Directors since October 2016, has in conjunction with the transaction resigned from the Cisco Board of Directors.

Cisco said the acquisition will enable it to take collaboration to the next level with even more intelligence. Accompany's AI technology and talent will help Cisco accelerate priority areas across its collaboration portfolio, such as providing user and company profile data in Webex meetings. Together, Cisco and Accompany will continue to power the future of work in a smarter way to enhance customer experiences.

"Amy has proven to be an effective and innovative leader through her years as an entrepreneur, an engineer, and CEO, and I couldn't be more pleased to have her and the Accompany team join Cisco," said Chuck Robbins, Cisco chairman and CEO. "Together, we have a tremendous opportunity to further enhance AI and machine learning capabilities in our collaboration portfolio and continue to create amazing collaboration experiences for customers."

"I am thrilled with the opportunity to join Cisco and the industry's leading collaboration team," said Amy Chang, Accompany founder and CEO. "Enterprise applications are rapidly becoming more intelligent and augmented with data and pertinent information in real-time. By combining Accompany's relationship intelligence capability with Cisco's award-winning collaboration product portfolio, customers will be able to more intelligently collaborate with employees, customers and partners."

In addition, Cisco announced that Rowan Trollope, current senior vice president and general manager of the Collaboration Technology Group, is leaving Cisco to become CEO at another company effective May 3.

In December 2016, Accompany raised $20 million in funding in a round led by Ignition Partners and participation from CRV. This brought total funding to $40 million.

Xtera's subsea repeaters enable C+L transmission on Seaborn's ARBR cable

Seaborn Networks will use Xtera’s wideband repeater to enable C+L band capability for its new ARBR cable between São Paulo and Buenos Aires.

The ARBR submarine fibre optic cable system, which is a fully-funded project developed jointly by Seaborn Networks and the Werthein Group. The 2,700 km open system, 4-fibre pair, 48Tbps, direct PoP-to-PoP subsea cable will connect Argentina and Brazil. The ARBR subsea cable system will allow for direct onward connectivity to New York, via the new Seabras-1 system.

Xtera’s wideband repeater is a hybrid Raman / EDFA design and can be configured to provide bandwidth in the C band alone, or across the C+L bands. First deployed in 2015, Xtera’s addition of Raman amplification to standard repeater technology has been used to achieve bandwidths of approximately 70 nm, while also offering very low noise solutions. Xtera said development continues to increase the capacity on a fibre pair to well over 100Tbit/s, further demonstrating that high capacity solutions do not have to mean large and costly fibre count systems.

In comparison, standard subsea repeaters use a single frequency band to carry the traffic. Seaborn said that by putting both the C and L frequency bands into ARBR’s cable on a fiber pair basis, it can pick and choose which fiber pair uses both bands and which use a single band, therefore enabling a choice between the most cost-efficient deployment of a fiber pair (single band) or an ultra-high capacity fiber pair (C+L). Seaborn is therefore able to tailor the subsea solution of each fiber pair to match the needs of different customers. For instance, carriers, ISPs and enterprise customers need affordable bandwidth from Argentina, whereas hyper-scalers need the assurance of ultra-high capacity throughout the life of the cable system.

“We work constantly with our partners and customers in this dynamic Latin American market to develop subsea cable systems that meet their future bandwidth demands,” says Larry Schwartz, Chairman & CEO of Seaborn. “Use of Xtera’s technology on the ARBR system will allow Seaborn to offer the most advanced system under the sea with on-demand capacities of up to 44Tbit/s per fiber pair. Disruptive innovation like this resonates with our content provider customers and positions us to respond to their needs well into the future.”

Xtera in conversation with Verizon - part 1



The evolution of optical transport technology in terrestrial and subsea networks as we approach Shannon's limit is the topic of discussion in this conversation with Glenn Wellbrock, Director, Backbone Network Design, Verizon, Stuart Barnes, Chairman and CSO, Xtera, and Vijay Rudravajjala, VP Engineering, Xtera.


See video:
https://youtu.be/H3zt-Nd-xQ8

Juniper sees better than expected Q1 results despite declining routing and switching sales

Juniper Networks reported net revenues of $1,082.6 million for the first quarter of 2018, a decrease of 11% year-over-year and 13% sequentially. GAAP operating margin was 5.1%, a decrease from 12.8% in the first quarter of 2017, and a decrease from 16.4% in the fourth quarter of 2017. GAAP net income was $34.4 million, a decrease of 68% year-over-year, resulting in diluted net income per share of $0.10. Non-GAAP net income was $99.5 million, a decrease of 44% year-over-year and 50% sequentially, resulting in diluted earnings per share of $0.28.

“We hit the high-end of our guidance during the March quarter due to better than expected results from our cloud vertical and another quarter of growth in our enterprise business," said Rami Rahim, chief executive officer, Juniper Networks. "We are encouraged by the trends we are seeing in several areas of our business and remain confident in our expectation to deliver sequential growth through 2018 and a return to year-over-year growth by the December quarter."

Some highlights from the company's quarterly report:

  • Cloud revenues were up slightly sequentially and ahead of the company's expectations. 
  • The Service Provider vertical was challenged due to the timing of customer deployments, resulting in decreases both year-over-year and sequentially.
  • Enterprise increased 4% year-over-year due to strength from all technologies. 
  • Routing product revenue amounted to $408 million, down 22% year-over-year and down 20% sequentially. 
  • Switching product revenue amounted to $230 million, down 5% year-over-year and down 1% sequentially. 
  • Security product revenue was $73 million, up 11% year-over-year and down 17% sequentially. 
  • Service revenue was $372 million, down 5% year-over-year and down 9% sequentially. 
  • Of the top 10 customers for the quarter, four were Cloud, four were Service Provider, and two were Enterprise. Of these customers, four were located outside of the U.S.
  • Sales in the Americas amounted to $588 million, down 17% year-over-year and down 17% sequentially. 
  • Sales in EMEA amounted to $308 million, up 8% year-over-year and down 5% sequentially. 
  • Sales in APAC amounted to $187 million, down 17% year-over-year and down 11% sequentially

MACOM reports upturn in its most recent fiscal quarter

MACOM reported revenue of $150.4 million for its fiscal second quarter ended March 30, 2018, a decrease of 19.2% compared to $186.1 million in the previous year fiscal second quarter and an increase of 14.9% compared to $130.9 million in the prior fiscal quarter. Gross margin was 43.6%, compared to 37.0% in the previous year fiscal second quarter and 46.6% in the prior fiscal quarter. Net loss from continuing operations was $15.5 million, or $0.50 loss per diluted share, compared to net loss from continuing operations of $134.3 million, or $2.21 loss per diluted share, in the previous year fiscal second quarter and net loss from continuing operations of $17.0 million, or $0.49 loss per diluted share, in the prior fiscal quarter.

“The December quarter marked the bottom of the cycle for MACOM in terms of revenue and demand, as evidenced by our 15% sequential growth. Across our served markets, order intake and customer forecasts returned to more normalized patterns in our fiscal second quarter,” commented John Croteau, President and CEO of MACOM.

“Following last year’s cyclical downturn in China, we believe we are entering the next phase of global infrastructure spending driven by 5G Telecom, continued strong investment by Cloud Service Providers, and now, a surge in defense spending and industrial capital investment. We’ve spent the last couple of years developing a portfolio of disruptive products and technologies to service these targeted areas of secular growth. Major customers have validated our technology and capabilities and are actively sponsoring us as we work to ramp volume.”

American Tower reports robust demand as sales grow 7.8% yoy

American Tower reported revenue of $1.742 billion for Q1 2018, up 7.8% over last year.

“The strong demand we experienced in late 2017 for our telecommunications real estate further accelerated in the U.S. as well as in our Latin America and EMEA regions in the first quarter of 2018. Notably, record levels of new business commencements, along with a robust pipeline of applications for both amendments and new colocations resulted in our increase in expectations for full year U.S. Organic Tenant Billings Growth to approximately 6.5% in 2018," stated Jim Taiclet, American Tower’s Chief Executive Officer.

During Q1, American Tower spent approximately $673 million to acquire nearly 10,600 sites primarily in international markets, including approximately 10,200 sites in India as part of its previously announced transaction with Vodafone India Limited.

Networking notes from Facebook's F8 Developer Summit

Here are some networking notes from Facebook's F8 developer conference this week in San Jose, California:

  • Facebook is fully invested in the Messenger platform.
  • Businesses are using Messenger to handle 8 billion customer messagers per month
  • Messenger is adding real-time translation between major languages
  • Facebook has attracted over 200,000 independent developers for Messenger
  • Facebook is adding Augmented Reality capabilities to Messenger and its bots
  • Over 300,000 bots have been created for Messenger
  • Facebook is also fully committed to WhatsApp, which has 450 million daily users.
  • WhatsApp is the largest implementation of end-to-end encryption
  • WhatsApp handles 65 billion messages per day, and these are not retained by the company.
  • WhatsApp is handling 2 million minutes of voice/Video calling per day
  • Group calling will soon be available on WhatsApp
  • Facebook is adding Augmented Reality features to Instagram.
  • Instagram is adding video chat.
  • Facebook began selling a standalone VR headset for $199.

Materials from the event are posted here: https://www.f8.com/

IDT launches High Baud Rate Linear Driver for 400G/600G

Integrated Device Technology (IDT) introduced its new GX76470 64G linear driver, in die form, for optical integrated modules, for 400G/600G coherent applications.

The driver is designed for OIF defined, highly integrated optical sub-assembly modules such as the HB-CDM (High Bandwidth Coherent Driver Modulator) and IC-TROSA (Integrated Coherent Transmitter-Receiver Optical Sub-Assembly) which enable miniaturization of optical transceiver modules and lowering the component cost for 400G ZR, metro, Data Center Interconnect (DCI) applications.  As such, the optical sub-assemblies are promising to be applicable to all the key small form factors: QSFP-DD, OSFP, CFP4-ACO, and CFP2-DCO.

"IDT's new GX76470 driver is another exciting addition to the expanding portfolio of data center and telecommunication solutions," said Dr. Koichi Murata, marketing director, Telecom, for IDT's Optical Interconnects Division. "Consumer and business demand for new, bandwidth-hungry applications and service like 5G, IoT, Smart City and virtual reality is driving the need for faster, more cost-effective data centers solutions that can be supported by our GX76470 driver and other new devices."

Interxion plans new data centres in AMS and Frankfurt

Interxion announced new data centre builds in Amsterdam (“AMS10”) and Frankfurt (“FRA14”), together with an expansion at the Science Park facility (“AMS9.2”) and the acquisition of land and a building at the Schiphol-Rijk campus in Amsterdam. The company is raising its 2018 annual capital expenditure guidance to €365 million - €390 million to account for the additional spending.
Tweet this

“Interxion is continuing to see a strong flow of opportunities across markets and customer segments and we are increasing our expansion programme to address this demand,” said David Ruberg, Interxion’s Chief Executive Officer. “Customers are recognising the value of our communities of interest and our trusted provider status for their mission critical applications. We are capturing deals from multiple customer segments, including connectivity, digital media, Cloud platforms and enterprises, and across the size spectrum. Of the announced phases in AMS9.2, AMS10 and FRA14, approximately 25% of the capacity is pre-sold.”

Monday, April 30, 2018

Verizon simplifies with multiservice edge

Verizon is using SDN to combine all of its existing service edge routers for Ethernet and IP-based services onto a single platform. Verizon is working with Cisco and Juniper on this new multi-service edge. The solution features a disaggregated control plane and leverages external compute to enhance the capabilities of that control plane beyond that of a traditional router.

“Software defined networking continues to deliver on its promise to improve network management and also enables us to be more nimble in the ways we serve our customers,” said Michael Altland, director, Network Infrastructure Planning, at Verizon. “By decoupling the control plane from a carrier-grade provider edge routing platform and moving it to general compute servers, we can serve our consumer and enterprise customers from the same platform, giving them all the functionality they need, while running our networks far more efficiently. This will also allow us to take advantage of future advances in server technology as our networks continue to grow.”

“Verizon continues to cross key milestones in transforming its networking practices to maximize performance and simplify operations,” said Sumeet Arora, senior vice president of engineering, Service Provider Business, Cisco. “With this new flexibility, Verizon can develop and launch innovative services for its customers faster, with improved efficiency.”

“Next-generation services that require low latency and real-time response are moving closer to users at the network edge, creating new gains in performance and business agility,” said Bikash Koley, chief technology officer, Juniper Networks.

Company profile: Adolite, an optical components start-up in Silicon Valley and Taiwan

Unlocking manufacturing bottlenecks for optical transceivers may be key in the race to 400G

Nearly two dozen companies announced 400G capabilities of some sort at the recent OFC 2018 conference, including transceivers in various formats, active optical cables, backplane interconnects, interface cards, optical module drivers, test equipment, and even full-blown switches. 

A few months ago, Broadcom announced commercial shipments of its StrataXGS Tomahawk 3 Ethernet switch silicon, boasting 12.8 Terabits/sec in a single device – enough to drive 32 x 400GbE ports. It has since followed up with the commercial shipment of a 400G gearbox device for hyperscale data centre and cloud applications – the BCM81724. This device is an 8x56-Gbps PAM-4 to 16x25-Gbps NRZ forward and reverse gearbox designed to enable next-generation high-performance switches with PAM-4 I/Os to connect to the large existing ecosystem of switches and plug-in modules with NRZ interface. We should see data centre switches with 400G ports on the market soon.

Put all of these together and we have a 400G ecosystem that is primed for rapid growth. Hyperscale data centres say their networks are besieged with a flood of east-west data flows. They are ready to deploy 400G backbones.

However, volume production of 400G transceivers may be a gating factor that holds back mass deployment of 400G data centre backbones for much of 2018 and into next year. Simply put, the market may remain supply constrained until transceiver manufacturers bring more manufacturing capacity online. This is difficult to do because building the highest performance optical transceivers requires skilled labor and specialised equipment to precisely align light sources, lenses and fibre in a repeatable fashion. The manufacturing, especially when we are talking about the multiple lanes required to achieve 400G, is hard to do.

Adolite, a privately-held start-up with its head office in Santa Clara, California and its manufacturing base in Taiwan’s Hsinchu Science Park, was founded earlier this year with a vision to solve this problem. The company has developed a breakthrough optical interconnect solution that simplifies the manufacturing of optical transceivers and on-board optics in high volume.

Adolite’s key innovation is to embed optical waveguides and electrical circuits into a single layer of flexible polymer circuit (FPC). The process directly integrates lasers and photo diodes onto the FPC using flip chip bonding techniques, eliminating the need for lenses and difficult fibre alignment and bonding, which is time consuming. Conventionally, microscopes were needed for the difficult and imprecise fibre alignment step and this led to low-yields and high costs in transceiver manufacturing.
By directly embedding the optical waveguides and electrical circuits into the FPC, the manufacturing process is greatly simplified and yields should go up, leading to faster production and lower costs. 
 
While other companies are using FPC technology, their implementations have been electric-only FPC bonded to optical layers, still requiring lenses and complex alignment during manufacturing.

Much of Adolite’s innovation is centred on the process of integrating optical reflectors and polymer waveguides on a single FPC layer. Adolite says thermal management and material science techniques enable its FPC to handle 400G data rates and up. The polymer material is sourced from Japan. The company says its design also uses significantly less power – perhaps as little as 10 percent of its competitors – which would also be a strategic advantage in dense data centres. Patents are pending. The company is also on track to receive ISO 9001: 2015 certification in the 2nd half of 2018 for its manufacturing operations in Taiwan. Patent filings are underway.

Adolite is using its technology to build its own line of optical transceivers and on-board optic solutions for 25G, 100G, 200G, 400G and upwards. It product plans extend from 25 SFP28  AOCs to 400G QSFP DD PAM4 (FR4) transceivers. Adolite expects to have volume production of its 400G solutions by Q1 2019.

For a start-up, ramping up from prototype to manufacturing in only twelve months is a challenge.  In this case, there will be big rewards for companies that open up the 400G market. Adolite is headed by Abraham Jou (CEO), who worked five years on the R&D team at Apple and went on to found two start-ups, PayEase (payments and big data processing) and Silicon Valley Communications (optical communications). Its technical team includes Dr. David Chung, CTO, Dr. Paul Wu in the critical role of EVP of Production, and Dr. Kenny Young as Principle Engineer.. The company has not disclosed its investor or its funding level to date, but no doubt will attract the attention of the venture capital community as its transceivers based on its FPC technology are put to the test.

The Race to 400G

Adolite’s simplified production process could be especially useful to hyperscale data centres operators who find a constrained market for transceivers. Broadcom may already be shipping its 400G silicon to hyperscale data centre operators who are designing and building custom switches for their backbones. Clearly, a very large number of 400G ports will be needed in data centres hosting 100s of thousands of Xeon servers with 25G interfaces. Adolite’s flexible polymer circuit is a promising solution to ramp up manufacturing