Showing posts with label OND Commentary. Show all posts
Showing posts with label OND Commentary. Show all posts

Tuesday, March 6, 2018

Eutelsat seeks new opportunities - part 2

In the first half of this article, we looked at the recent financial performance of Eutelsat, the Paris-based satellite operator known for its pioneering video distribution services and global reach. Currently, the Eutelsat fleet of satellites is carrying 6,810 TV channels, of which 1,275 channels are HD – a penetration rate of only 19%, which seems low in this era when 4K television screens are now widely available in many markets at declining prices.

As the video market continues to shift toward over-the-top services, and as fibre reaches many of the locations which were previously the exclusive domain of satellites, Eutelsat will seek new opportunities, such as in-flight mobile connectivity -- its only segment that is currently growing. Eutelsat also has interesting ventures underway in Africa, and recently in the "Belt and Road" countries in partnership with China Unicom.

Upcoming satellite launches

Unlike some of its industry rivals that have drawn up plans for dozens, hundreds or even thousands of new satellites in the coming decade, Eutelsat seems to be taking a conservative CAPEX approach with regards to the orders book. While current shareholders may appreciate keeping a healthy dividend, the space business is inherently a costly one. Eventually, satellites get old and must be replaced. Newer technology means that the current assets in orbit can depreciate quite quickly.

As in the case of current satellites not optimized for 100% HD or even 4K video. There will also be the need to defend against the ultra-aggressive business plans of SpaceX and others, who could conceivably push an established company such as Eutelsat into irrelevancy quite quickly. Here is a list of upcoming Eutelsat satellite launches:

A Vision for Africa

In September 2016, Eutelsat suffered a major setback in its vision to deliver broadband connectivity to Africa in partnership with Facebook when SpaceX experienced a launch pad anomaly during a pre-launch test at Cape Canaveral, Florida, leading to the explosion of a Falcon 9 rocket and the destruction of its payload, Eutelsat's AMOS-6 satellite.

AMOS-6 was a Ka-band geostationary satellite configured with high gain spot beams for covering large parts of West, EAast and Southern Africa. The satellite was designed for community and Direct-to-User Internet access using affordable, off-the-shelf customer equipment.
The partnership with Facebook was announced in October 2015 with a mission to reach large parts of sub-Saharan Africa.

Eutelsat’s upcoming African Broadband Satellite looks to be a replacement for this lost bird. The new satellite t has ordered from Thales Alenia Space for possible launch in 2019. The all-electric satellite will be the first to use Thales Alenia Space's new Spacebus Neo platform. And it will offer 75 Gbps of capacity across a network of 65 spotbeams, which together provide quasi-complete coverage of Sub-Saharan Africa. The satellite will address direct-to-user consumer and enterprise broadband services using dishes from approximately 75 cm.  Villages could use the satellite connectivity for Wi-Fi hotspots and mobile phone backhauling.  No word on whether Facebook will play a role in the renewed project.

There are two other past regional expansions for Eutelsat that should be noted. In October 2017, Eutelsat acquired Noorsat, one of the leading satellite service providers in the Middle East, from Bahrain’s Orbit Holding Group. In 2013, Eutelsat acquired Satélites Mexicanos, for $831 million in cash plus assumption of $311 million in Satmex debt. Satmex operates three satellites that cover 90% of the population of the Americas.

Finally, Eutelsat and China Unicom signed a memorandum of understanding to address satellite communications market in the framework of China’s “Belt and Road” initiative. The agreement builds on in-flight connectivity deal formed between Eutelsat and UnicomAirNet (UAN), which was recently formed by China Unicom's broadband network unit and Hangmei, a Chinese Wi-Fi service and content provider for railways and buses, to provide IFC services to Chinese commercial airlines.

The Belt and Road Initiative is the development strategy proposed by the Chinese government that focuses on connectivity and cooperation between Eurasian countries. The idea is to build a 21st century Silk Road linking more than 68 countries, equivalent to 65% of the world's population and 40% of the global GDP as of 2017. The Belt and Road Initiative has the support of President Xi and the highest levels of his administration, leading some analysts to call it the largest organised development in history.

China's leading IT players, including China Telecom, China Unicom, and China Mobile, are being encouraged to pursue an overseas expansion strategy. As such, Eutelsat makes an interesting partner for China Unicom in that it already has the satellite coverage, and the local TV broadcast partners), in the regions that China Unicom wishes to expand. Likewise, if Eutelsat is looking for a partner with the financial resources, scale, and ambition to cover Asia, the Middle East and Africa, China Unicom presents many possibilities. Global politics, of course, would play a role, with government contracts and national security interests as top concerns if this partnership were to deepen.

Monday, February 26, 2018

All data centre traffic becomes cloud data centre traffic

Cisco is now predicting that global cloud data center traffic will reach 19.5 zettabytes (ZB) per year by 2021, up from 6.0 ZB per year in 2016 – a 3.3-fold growth, representing a 27 percent compound annual growth rate from 2016 to 2021.  By that year, the Cisco forecasters will have a very difficult time distinguishing regular data centre traffic from cloud data centre traffic.  Fully 95 percent of data centre traffic will be cloud data centre traffic.  This does not necessarily mean public data centre traffic, just that nearly all data centres, public and private, will have adopted cloud virtualisation technologies by that date. Cisco’s definition of cloud encompasses virtualisation in networks, servers, storage, applications, and services.

Earlier security concerns about sharing servers and storage resources amongst applications, even within an organisation, have given way to the forces of cost and power efficiency. Data is now much less stored and processed in a confined physical environment. More and more, even “data at rest” moves to wherever the algorithms of efficiency demand.  This has huge implications not only for large data sets moving across wide-area boundaries, but also for east-west traffic within data centre campuses.  Virtualisation means that data forever will be on the move.

The newly-published Cisco Global Cloud Index (2016-2021), which is now in its seventh annual edition, predicts that the number of hyperscale, public cloud data centres will nearly double from 338 in 2016 to 628 globally in 2021.  It will take a massive construction effort to pull this off.  Nearly every month, we report when AWS, Facebook, Google, Microsoft, Apple, Alibaba, IBM, Oracle etc. unveils plans for new facilities. Typically, these are warehouse-sized builds on a new plot of land in a remote location, where renewable energy can be procured in quantity and at a reasonable cost. Increasingly, we are seeing these data centre campuses being built close to urban centres.
With this level of expansion, one wonders why certain telcos (Verizon, Centurylink and possible AT&T) are selling off their data centers rather than holding them as strategic assets or appreciating investment. Possbily, these facilities are too old and would required extensive HVAC upgrades to accommodate the high number and density of servers that hyperscale cloud facilities require. Or maybe they realize that simply cannot compete with the likes of AWS or Microsoft Azure, so better to exit the business sooner rather than later. It is odd given the surge in cloud data centre traffic that Cisco is predicting.

Some other key findings - by 2021, hyperscale data centres will support:

  • • 53 percent of all data centre servers (27 percent in 2016)
  • • 69 percent of all data centre processing power (41 percent in 2016)
  • • 65 percent of all data stored in data centres (51 percent in 2016)
  • • 55 percent of all data centre traffic (39 percent in 2016)
  • • By 2021, 94 percent of workloads and compute instances will be processed by cloud data centres; 6 percent will be processed by traditional data centres.
  • • Overall data centre workloads and compute instances will more than double (2.3-fold) from 2016 to 2021; however, cloud workloads and compute instances will nearly triple (2.7-fold) over the same period.
  • • The workload and compute instance density for cloud data centres was 8.8 in 2016 and will grow to 13.2 by 2021. Comparatively, for traditional data centres, workload and compute instance density was 2.4 in 2016 and will grow to 3.8 by 2021.
  • • Globally, the data stored in data centres will nearly quintuple by 2021 to reach 1.3 ZB by 2021, up 4.6-fold (a CAGR of 36%) from 286 EB in 2016.
  • • Big data will reach 403 exabytes (EB) by 2021, up almost 8-fold from 25 EB in 2016. Big data will represent 30 percent of data stored in data centres by 2021, up from 18 percent in 2016.
  • • The amount of data stored on devices will be 4.5 times higher than data stored in data centres, at 5.9 ZB by 2021.
  • • Driven largely by IoT, the total amount of data created (and not necessarily stored) by any device will reach 847 ZB per year by 2021, up from 218 ZB per year in 2016. Data created is two orders of magnitude higher than
data stored. 

  • • By 2021, big data will account for 20 percent (2.5 ZB annual, 209 EB monthly) of traffic within data centres, compared to 12 percent (593 EB annual, 49 EB monthly) in 2016.
  • • By 2021, video streaming will account for 10 percent of traffic within data centres, compared to 9 percent in 2016.
  • • By 2021, video will account for 85 percent of traffic from data centres to end users, compared to 78 percent in 2016.
  • • By 2021, search will account for 20 percent of traffic within data centres by 2021, compared to 28 percent in 2016.
  • • By 2021, social networking will account for 22 percent of traffic within data centres, compared to 20 percent in 2016.
  • •  By 2021, 75 percent (402 million) of the total cloud workloads and compute instances will be SaaS workloads and compute instances, up from 71 percent (141 million) in 2016. (23% CAGR from 2016 to 2021).
  • • By 2021, 16 percent (85 million) of the total cloud workloads and compute instances will be IaaS workloads and compute instances, down from 21 percent (42 million) in 2016. (15% CAGR from 2016 to 2021).
  • • By 2021, 9 percent (46 million) of the total cloud workloads and compute instances will be PaaS workloads and compute instances, up from 8% (16 million) in 2016. (23% CAGR from 2016 to 2021).

Thursday, February 8, 2018

New market highs for semiconductor sales

The first financial reports of 2018 from semiconductor conductor companies have started to arrive. Intel reported very strong results for the fourth quarter of 2017, as did Microsemi (see below). The figures are encouraging for the networking and telecommunications businesses as well given that nearly every system-on-chip solution ends up in a device that is network connected. The more end-user devices, the more nodes on the network, and the data tsunami continues to grow.

Last week, Gartner updated its forecast of worldwide semiconductor revenue, predicting a total $451 billion in 2018, an increase of 7.5 percent from $419 billion in 2017, The new figure is nearly double the 4 percent growth rate that the firm had predicted earlier. The reason for the increased optimism are “more favourable market conditions”, especially for DRAM and NAND memory. Gartner sees the possibility of price increases for some semiconductor categories during 2018, which in turn would put pressure on margins for system vendors and smartphone manufacturers.

Gartner is predicting other categories will grow at a 4.6% clip in 2018, including field-programmable gate array (FPGA), optoelectronics, application-specific integrated circuits (ASICs), nonoptical sensors and application-specific standard products (ASSPs). This analysis comes from Gartner’s newly published  "Forecast Analysis: Electronics and Semiconductors, Worldwide, 4Q17 Update.

"Favorable market conditions for memory sectors that gained momentum in the second half of 2016 prevailed through 2017 and look set to continue in 2018, providing a significant boost to semiconductor revenue," said Ben Lee, principal research analyst at Gartner. "Gartner has increased the outlook for 2018 by $23.6 billion compared with the previous forecast, of which the memory market accounts for $19.5 billion. Price increases for both DRAM and NAND flash memory are raising the outlook for the overall semiconductor market."

This momentum has also been reported by The Semiconductor Industry Association (SIA) which found that worldwide sales of semiconductors reached $37.1 billion for the month of October 2017, an increase of 21.9 percent from the October 2016 total of $30.4 billion and 3.2 percent more than September’s total of $36.0 billion.  Simply put, October was global semiconductor industry’s largest-ever monthly sales total.
“The global semiconductor market continued to grow impressively in October, with sales surpassing the industry’s highest-ever monthly total and moving closer to topping $400 billion for 2017,” said John Neuffer, president and CEO, Semiconductor Industry Association. “Market growth continues to be driven in part by high demand for memory products, but combined sales of all other semiconductor products were up substantially as well, showing the breadth of the market’s strength this year.”

Intel builds on its data centre strength

Despite the PC market continuing its long-term decline due to consumers not replacing traditional home PCs, Intel is telling the financial community that 2018 will be another record year. Its earnings report highlights its Data Center Group as the leading growth driver for Q4 2017.
DCG's revenue was up 20%. This breaks down as follows: cloud segment was up 35%, communications service provider revenues were up 16%, enterprise was up 11%, and adjacency revenue was up 35%. Overall unit volume was up 10%. The new Xeon Scalable server processors, which were launched in July,  are ramping well.

With all of the new data centre construction underway pretty much in every major metro, it is easy to see how Intel’s cloud sales were up so strongly. Many are wondering if the recently disclosed security vulnerabilities in Intel CPUs will hurt its business.  With the (buggy) firmware patches seen as only a temporary fix, there is a distinct possibility that the operators of cloud data centres will choose to retire the current crop of processors earlier than expected once redesigned silicon hits the market. In that case, the refresh cycle for CPUs in cloud data centres could be brought forward. Intel will still be the leading supplier, even if customers are annoyed or infuriated by the silicon bug, and this would benefit Intel financially rather than hurt it. It’s too early to say this will play out, and industry forecasts have no data at this point.

Growing prospects for semiconductors in 2018 forward
Data centres
5G infrastructure
Bluetooth and Wi-Fi
Automotive segment
Ethernet ICs
AI, machine learning, and machine vision

Tuesday, February 6, 2018

Chasing the next virtual network opportunity

by James E. Carroll

For tennis fans, all attention is currently on the city of Melbourne, where the final rounds of the Australia Open are underway. For those interested in the future of cloud connectivity, the focus goes to the city of Brisbane, where an Australian upstart is making a name for itself in the emerging Network-as-a-Service (NaaS) category. Potential users of NaaS could include large multinational, governments, other cloud providers. service providers including mobile network operators, and mobile virtual network operators (MVNOs). We expect to see many mobile operators transition to virtual networks over time and this could be one model.

Founded in 2013, Brisbane-based Megaport has built what it considers to be the world's first SDN interconnection fabric linking enterprises with equipment in colocation data centres to leading cloud service providers.

Megaport says its elastic fabric is "the reason cloud connectivity will scale." The company developed and runs its own proprietary software stack for automating virtual connections across the fabric. It offers APIs that would enable customers to automate connections across its service. Megaport owns and operates its own core network infrastructure, including fibre and transport for each market and between inter-city data centres. This footprint now covers major cities in Australia, Asia Pacific, North America, and Europe – a total of 37 major markets in 19 countries. Unlike with best-effort public Internet access, Megaport is able to provide strict SLAs because it controls the transport network and the switching fabric, and the Layer 2 access inside the colo data centre.

The ASX publicly-listed company has reported steady revenue growth over the past year as it quickly scales its service. For its fiscal quarter ending 31-December-2017, revenue was A$4.68 million, up 12.7% sequentially. The total number of customer ports increased in the quarter to 2,259, up 9% sequentially.

From Virtual Layer 2 to Virtual Layer 3

Just this week, Megaport is unveiling a virtual router service that enables customers to rapidly and privately connect at Layer 3 without the need to own or manage routers or physical infrastructure.  The Megaport Cloud Router (MCR), which rides the company’s same physical network, aims to make it easier for companies to expand their service footprint through virtual Points of Presence (PoPs), and peer with ecosystem partners worldwide. It does so by removing the need to own physical routers or network infrastructure. Megaport said its service also enables cloud to cloud connectivity. Customers can use its cloud router to move workloads and data between Cloud Service Provider (CSP) environments.  Customers can create virtual routers within routing zones around the world to enable global coverage and support localized routing decisions. In addition, networks service providers connect to Megaport can use MCR to set up virtual PoPs around the world.

“As a Network as a Service company, it’s imperative that Megaport continues to innovate solutions that abstract complexities in the network buying experience,” said Vincent English, Chief Executive Officer, Megaport. We’ve moved further up the stack by expanding our SDN’s capabilities to address Layer 3 IP routing and support a broader set of customers with varying technical capabilities and business needs. With Megaport Cloud Router, there’s no need for a deep understanding of Layer 3 intricacies to take advantage of IP routing features. Cloud to cloud connectivity is one of several new use cases unlocked by MCR which provides powerful options for enterprises architecting next-generation multicloud and hybrid cloud solutions. Our customers can move beyond the constraints of their physical network and rapidly establish virtual Points of Presence to unlock unique peering and interconnection opportunities around the world.

Company leadership

Megaport was founded by Bevan Slattery, who currently serves as Chairman of the business. Over his career, Slattery built multiple successful Australian IT and telecommunications companies including Superloop and NextDC.  He also co-founded PIPE Networks which grew to become Australia’s largest Internet Exchange and Australia’s third largest metropolitan fibre network provider, selling to TPG in May 2010.

Megaport is headed by Vincent English, who previously was Chief Financial Officer for Digicel Group. Prior to joining Megaport, Vincent was for Digicel Group, the global mobile network operator active in  31 markets in the Caribbean and South Pacific.  Megaport’s engineering team is led by Tim Hoffman (CTO), who previously led the Global Network team at Twitter, responsible for worldwide infrastructure, including all interconnection, backbone and content distribution infrastructure, and global data centres.  In this role, Hoffman negotiated peering agreements with some of the largest Internet backbone providers. Eric Troyer serves Megaport's Chief Marketing Officer. Troyer previously was Director of Network Edge and Interconnection Strategy at Microsoft where he led planning and engineering teams tasked with network expansion and IP capacity acquisition to scale Microsoft’s cloud strategy. Working at Equinix, he drove the Equinix Internet Exchange product

Building the Ecosystem

Customers connect to its fabric via a single, physical "mega-port" at any of the 185 colo data centres in which is present. This physical port enables the set-up and tear-down of virtual ports to any of the other parties connected to the global Megaport network. The concept is simple. Once a sufficient number of parties are on-board, Megaport benefits from the "n-squared" magic of networks. The company is now poised to enter that rapid growth phase.

Megaport has been prolific in forming partnerships with key players for cloud. These can be sorted into the following categories:
  • Cloud Service Providers: Alibaba, AWS, Microsoft, Google Cloud, Oracle Cloud
  • Data Centre Operators: now present in 185 data centres worldwide, including those of CyrusOne, Digital Realty, EdgeConnex, 4 Degree Data Centres, IO, Cyxtera, vXchnge, QTS, FORTRUST, Stream Data Centers 
  • Network and Managed Service Providers: Aqua Comms, Cloudlogix, GT, Seaborn Networks, Rackspace

In the last category of network service providers, we see two subsea cable operators. This is interesting because it means that enterprises attached to the Megaport fabric now have the ability to activate transoceanic capacity on a short-term basis and via a simple web portal. The partnership with Aqua Comms, which was announced in November 2016, allows customers to turn on elastic interconnectivity services to Aqua Comms’ transatlantic subsea network between New York, Dublin, and London.  Consumption is be based on cloud computing models, including month-to-month services – a huge gain in provisioning flexibility compared with the old system of negotiating 20-year contracts (IRUs).

In its home country of Australia, which is probably its most developed market, Megaport is already providing a similar capability to the U.S.  Dedicated capacity between Sydney and Los Angeles now enables its enterprise customers in Australia and New Zealand to connect to multi-national cloud nodes in North America.

Reliance Jio continues its rapid rise

by James E. Carroll

How fast can a network grow from zero to 160 million subscribers? Ask Reliance Jio, the brainchild of billionaire investor/entrepreneur/tycoon Mukesh Ambani and his Reliance Industries Ltd.
Reliance Jio only first launched commercial service on 5th September 2016.

Its market debut has been described as an earthquake for the Indian telecoms market, and the start of a price war that is drawing the casualties from the nation’s twelve mobile operators. In addition to cut-rate tariffs and the promise of unlimited LTE mobile data for an extended period, the company tapped into the energy and graces of Prime Minister Narendra Modi. Chairman Ambani even dedicated the new company to “”realising the Prime Minister’s inspiring vision of Digital India for 1.2 billion Indians. Despite this connection, Jio immediately drew complaints from other operators for what they saw as anticompetitive behaviour.

From the starting line on 05 September 2016, Jio rocketed ahead to become the fastest growing mobile operator ever seen. In the first month, Jio enrolled 16 million lines. The 50 million threshold was passed on the 83rd day. The 100 million milestone came on 22 February 2017. As of 31 December 2017, Jio passed the 160 million subscriber milestone, adding 27 million gross user lines in the fourth quarter alone.

As the Indian market tends to experience greater churn rates and a higher percentage of prepaid users, Jio’s net additions for the quarter amounted to 21.5 million.  In Q3 2017, the company enrolled 19.5 million lines, indicating that Jio’s remarkable clip continues.

Six months after launch, the Telecom Regulatory Authority of India (TRAI) ordered Jio to withdraw the 3 months complimentary, unlimited mobile broadband offer for new subscribers, arguing that such a generous offer distorted the market.

This week, Reliance Jio is reporting its second quarterly profit in its brief history and even as it continues to spend aggressively to build out its network.

For its most recently fiscal quarter, Reliance Jio reported standalone revenue from operations of  6,879 crore  rupees  (US$1.7 billion), and up 11.9% over trailing quarter. Standalone EBITDA  amounted to 2,628 crore rupees (US$411.8 million) and the EBITDA margin was 38.2% (trailing quarter at 23.5%).  Standalone net profit amounted to 504 crore rupees (US$78.9 million).
Before we become too enamored of this operator, it is critical to note the very low ARPU levels in the Indian telecoms sectors compared to those in developed economies.

Jio’s average revenue per user per month is just154 rupees (approximately $2.41)  

At these levels, continued investment in the latest generation of imported networking equipment will require commitment from the parent firm, Reliance Industries, or the emergence of adjacent opportunities, such as mobile banking or shopping services that could be commercialized with the very large subscriber base.

Additional metrics disclosed by Jio
  • World’s largest mobile data consumption network – first Exabyte network in the world
  • Total wireless data traffic of 43,100,00,000 GB (9.6 GB per subscriber per month) 
  • Total voice traffic of 311,130,000,000 minutes
  • Video consumption has crossed 2,000,000,000 hours per month on the network (13.4 hours of video consumption per subscriber per month)
  • On track to achieve 99% population coverage during the year
  • Only network to deploy pan-India 4G across the 800MHz/ 1800MHz/ 2300MHz bands 
Acquiring more infrastructure

Earlier this month, Jio agreed to acquire network infrastructure assets of Reliance Communications Limited and its affiliates. Jio was the winning bidder in a sale mandated by the lenders of Reliance Communications.

The sale includes assets under four categories – Towers, Optic Fiber Cable Network, Spectrum and Media Convergence Nodes, specifically:

122.4 MHz of 4G Spectrum in the 800/900/1800/2100 MHz bands
Over 43,000 towers, amongst the top 3 independent tower holdings in India
~ 1,78,000 RKM of fiber with pan India footprint
248 Media Convergence Nodes, covering ~5 Million sqft used for hosting telecom infrastructure

The deal was valued at US$$3.77 billion, according to media reports. Reliance Communications said it will use the proceeds for debt repayment and that it retains its other businesses including its enterprise networking practice, its data centers, and its subsea cable network.

Jio’s management said the assets are strategic in nature and are expected to contribute significantly to the large-scale roll-out of wireless and Fiber to Home and Enterprise services in India.

Here’s what we know about Jio’s physical network. 

The Jio All-IP digital platform is built on Cisco’s Open Network Architecture and Cloud Scale Networking technologies featuring IP/MPLS, spanning areas including Data Center, Wi-Fi, Security and Contact Center solutions. Jio has laid more than 185,000 miles (or 300,000 KM) of fiber, and built India’s largest cloud data center to build platforms for applications and vertical solutions. Cisco claims a leading role at this layer of the network.

Nokia provided optical core and metro solution for Reliance Jio Infocomm's (Jio) pan-India 4G LTE network to support traffic growth created by the operator's initiative to deliver broadband connectivity for all of India.  As part of this deployment, Nokia is providing a 100 Gbit/s transport network that spans 90,000 km designed to enable Jio to offer high-capacity broadband services to underserved regions throughout India, as well as support nationwide long-distance (NLD) service.

In March 2017, Ericsson announced that it was providing its OSS fulfilment suite as part of Jio's broadband network deployment.

Monday, February 5, 2018

Transpacific subsea capacity on the rise

We’ve entered a period in which subsea capacity along the major transoceanic routes is expanding by leaps and bounds, and the Pacific is no exception.

For years, the four fibre pair Pacific Crossing (PC-1) has ruled the roost for bandwidth between the west coast of the United States and Asia. Pacific Crossing's 21,000 km fiber optic ring first entered service in 2001 as a protected system with a capacity of 3.2 Tbps based on its ability to carry 10 Gbps wavelengths. A mid-life upgrade enabled the ability to carry 100G wavelengths, effectively tripling capacity to 10 Tbps.

In 2009, NTT acquired the cable system from Pacific Crossing Ltd, Inc., a former subsidiary of Global Crossing and of the former Asia Global Crossing. During the great Tohoku earthquake and tsunami of March 2011, the PC-1 North and PC-1 West segments were severed underwater, but were subsequently restored a few weeks later by the cable repair ship, CS Lodbrog. Service on the southern portion of PC-1 was unaffected by the disaster.

The 17-year-old PC-1 subsea cable network is now showing its age.  The new generation of cables offer much greater capacity and, some, even the ability to carry future 400G wavelengths. We are also seeing

The FASTER transpacific cable

The FASTER transpacific cable entered commercial service in June 2016 – a milestone event described at the time as “the world's highest capacity undersea cable system.”  Google touted the fact that it controls a single fibre pair on the 9,000km trans-Pacific cable connecting Oregon and two landing sites in Japan (Chiba and Mie prefectures).  Over this single fibre pair, Google is able to carry up to 100 wavelengths at 100 Gbps – the equivalent of 10 Tbps of traffic when fully loaded.  The design features extremely low-loss fiber, without a dispersion compensation section, and the latest digital signal processing technology. NEC was the lead contractor.  Significantly, it was the first major transpacific cable system to land in Japan after the Tohoku mega-earthquake and tsunami of 2011.

JUPITER will rule (at least for a while)

In October 2017, plans were unveiled for JUPITER, a new large-capacity, low-latency subsea cable between Japan and the United States with the backing of SoftBank, Facebook, Amazon, PLDT and PCCW Global.

The JUPITER cable system. which will have a total length of 14,000 km, will have two landing points in Japan — the Shima Landing Station in Mie Prefecture and the Maruyama Landing Station in Chiba Prefecture — as well as a U.S. landing station in Los Angeles, California, as well as a landing station at Daet in the Philippines.

 JUPITER will feature a state-of-the-art submersible ROADM employing WSS (wavelength selective switch) for a gridless and flexible bandwidth configuration. The cable system will also be designed to support 400 Gbps wavelengths. The initial design capacity is 60 Tbps.

NTT Com said its Asia Submarine-cable Express (ASE), Asia Pacific Gateway (APG) and Pacific Crossing-1 (PC-1) cables will connect with JUPITER to provide a redundant three-route structure linking major cities in Asia, Japan and the United States. NTT Com is also planning direct connections from the cable landing stations in Japan to data centers in Tokyo and Osaka.

JUPITER is expected to come online in early 2020

Telstra expands its transpacific subsea cable investments

Telstra has just announced plans to invest in two transpacific cable projects, expanding its transoceanic network by the equivalent of 6 terabits per second of new capacity. Most of Telstra’s transpacific traffic is currently carried on the AAG cable, which connects South East Asia to the US west coast via Hong Kong, Guam and Hawaii. The 20,000 Km AAG system went in service in late 2009. The system features 96*10G DWDM technology. In addition to Telstra, there are 18 other carriers participating in the consortium.

The first new Telstra undersea investment is a newly announced, 13,000-km Hong Kong Americas (HKA) cable project, which is also backed by China Telecom, China Unicom, Facebook, and Tata Communications. The HKA system will feature six fiber pairs and will connect from Chung Hom Kok in Hong Kong to Hermosa Beach in California. The designers are considering additional branching segments. It’s design capacity is stated at 80 Tbps. Alcatel Submarine Networks (ASN), which is the lead contractor for the project, will supply its submarine WSS ROADM units and the latest generation of repeaters.  In addition, the HKA cable will be compatible with future generations of submarine line terminal equipped with Probabilistic Shaping technology. Telstra will own a half-fibre pair on the HKA cable system.

The second Telstra project is the 12,800-km Pacific Light Cable Network (PLCN), which will also connect Hong Kong and the U.S. and offer an estimated cable capacity of 120 Tbps.  The cable is expected to enter service in the summer of 2018. This project is organized by Pacific Light Data Communication Co. Ltd., a new company based in Hong Kong, and has the backing of Google and Facebook. TE SubCom has been appointed lead contractor.

Carrier-neutral Hawaiki will be ready for service in June

The deployment of Hawaiki, a new carrier-neutral submarine cable linking Australia, New Zealand, Hawaii and Oregon, is progressing quickly as the new fibre is laid across the vast South Pacific.  This 15,000 km undersea system promises up to 43 Tbps of new capacity on the market. The project is owned by Hawaiki Submarine Cable LP, headquartered in Auckland, New Zealand. The key people behind the project were New Zealand-based entrepreneurs Sir Eion Edgar, Malcolm Dick and Remi Galasso.

TE SubCom serves as the lead contractor

More than half of the 15,000 route is now in the water and landing stations in Oregon, Hawaii and Sydney  have been built. Another cable landing station in American Samoa will be ready in March. TE SubCom’s cable-laying vessel CS Responder is now berthed in Auckland, poised to begin marine activities for the New Zealand leg of the transoceanic cable system later this month. The operation will include the landing of the Hawaiki cable in Mangawhai Heads. With FCC licensing now complete, this week the company is predicting that the whole system can be declared commercially ready-for-service in June.

“The start of 2018 finds Hawaiki closer and closer to ready for service”, stated Remi Galasso, CEO of Hawaiki. “Landing the cable in its home country represents a major event for our team and I would like to take this opportunity to thank all our New Zealand partners for their continuous support. Hawaiki will bring huge benefits to New Zealand in terms of greater connectivity to Australia and the US, security of supply, diversity and increased business opportunities for the Telecom and IT industries.”

Monday, January 29, 2018

New cables add to transatlantic subsea capacity essential to operators

by James E. Carroll

Much new and needed subsea capacity is going into Atlantic waters this year. This new capacity will be essential to Internet Content Providers, public cloud companies, and mobile operators as they push into 5G.

2018 opened with the unveiling of HAFVRUE (mermaid in Danish), a massive subsea cable project that will link New Jersey to the Jutland Peninsula of Denmark with a branch landing in County Mayo, Ireland. Optional branch extensions to Northern and Southern Norway are also included in the design.

The Mermaid cable has a theoretical design capacity of 108 Tbps, which will make it one of the high-capacity subsea cables ever built. TE Subcom has been signed as the system supplier for HAVFRUE. The construction contract is now in force and the marine survey is underway. A ready-for-service date is promised in Q4 2019 – less than 24 months away. The HAVFRUE subsea cable system will be optimized for coherent transmission and will offer a cross-sectional cable capacity of 108Tbps, scalable to higher capacities utilizing future generation SLTE technology.

What is most interesting about the Mermaid project is the diverse membership of the consortium, especially the inclusion of Facebook, Aqua Comms and Bulk Infrastructure as key members.

Perhaps it's not too surprising to see Facebook on this list, even though they must be one of biggest owners of transatlantic bandwidth at the moment. After all, Facebook, along with Microsoft, is co-owner or the newly commissioned MAREA cable system, which spans 6,600 km from Virginia Beach, Virginia to Bilbao, Spain.  The MAREA cable has a record 160 Tbps design capacity using eight fibre pairs.  However, MAREA takes a more southern route than other transatlantic cables, which mostly connect northern Europe to the New York/New Jersey region.  MAREA’s Virginia landing makes a good connection point to Facebook’s 160-acre data centre campus in Forest City, North Carolina.  Perhaps there are plans for a Facebook data centre in southern Europe, given that the company’s current European facilities are all north. The social media giant famously activated its first European data centre in in 2013 in Lulea, a city on the coast of northern Sweden, where the sub-Arctic climate and cheap hydroelectricity were seen as especially advantageous. In January 2016, Facebook selected Clonee (a suburb of Dublin), Ireland as the location for its second data centre in Europe. (Facebook’s international headquarter has been in Ireland since 2009).  One year later, in January 2017, Facebook announced its selection of Odense, Denmark as the location for its third European data centre. The company said Denmark was chosen for its robust Nordic electric grid, access to fiber, access to renewable power, and a great set of collaborative community partners. Renewal energy is expected to account for 100% of electricity needs.

Given these two new data centres in Ireland and Denmark, it makes sense that Facebook would invest in a new transatlantic cable landing in these two countries.

For Aqua Comms, which is a young, subsea cable operator based in Dublin, the new HAVFRUE cable also adds to its growing existing transatlantic portfolio. It was only two years ago, in January 2016, that Aqua Comms’ AEConnect Cable System went into operation. The AEConnect cable spans more than 5,400 km across the Atlantic between Long Island, NY and  Killala, County Mayo, on the west coast of Ireland.  Its design capacity is 130 Tbps, or 130 wavelength services at 100Gbps per fibre pair. Aqua Comms’ first cable project was CeltixConnect, a 72-fibre pair subsea cable connecting Ireland and the UK that entered service in January 2012.

With this new HAVFRUE cable, Aqua Comms has been appointed system operator and landing party in the U.S., Ireland, and Denmark. The company plans market and sell capacity services and raw spectrum on its portion of the HAVFRUE cable system under the brand name America Europe Connect-2 (AEC-2).  Aqua Comms will also commission a new cable route to Denmark through the UK, developing CeltixConnect-2, which connects Dublin to Manchester, as well as the North Sea Connect cable that will link Stellium’s data centre in Newcastle, England, to Esbjerg, Denmark.
The new HAVFRUE cable not only adds capacity for Aqua Comms, it also enables the company to create a resilient, ring-based infrastructure between the East Coast of the U.S., Ireland, and Northern Europe that is especially attuned to hyperscale cloud companies needing dozens of 100G transatlantic circuits in the years ahead.

A subsea alliance for Europe to South America

In the south Atlantic, Seaborne Networks has been building a new generation of subsea cables. In September 2017, Seaborn activated its Seabras-1 direct subsea system between New York and São Paulo while bypassing the hurricane-prone areas of Florida, the Caribbean and Bermuda.. The new Seabras-1 submarine cable, which spans 10,600-km,, has multiple branching units and is designed to provide additional route diversity to Virginia Beach, Miami, St. Croix, Fortaleza, and Rio de Janeiro. Seaborne is planning a new direct subsea system between Brazil - Argentina (RFS Q4 2018); and SABR, a new subsea system between Cape Town, South Africa and Seabras-1 (RFS 2019).

Interestingly, Seaborn and Aqua Comms have just announced a strategic alliance to provide subsea connectivity between South America and Europe.  This looks to integrate Seaborn’s Seabras-1 subsea cable system now directly interconnects with Aqua Comms’ America-Europe Connect (AEConnect) subsea cable network. The two submarine cable systems will interconnect in Secaucus, New Jersey, in the location of Seaborn’s primary network operations centre.

Thursday, January 25, 2018

Cable Mergers and Spinoffs - Bigger is Better?

Nearly 15 months have passed since AT&T and Time Warner announced their $109 billion-dollar merger agreement. For most of 2017, the companies were confident that their merger would pass regulatory review by the Department of Justice and by the FCC. As the first big to face scrutiny from the income Trump administration, the presumption was that regulators would take a pro-business, hands-off approach especially since the companies do not compete in the same markets and hence would not be constricting the competitive field.  The predicted completion date was “by the end of 2017.” The deadline has now passed.  The new target is “by mid-2018.” 

What’s the hold-up? In late November, the U.S. Department of Justice filed a legal case to block the proposed AT&T + Time Warner merger, apparently on the grounds that the size of the combined company will but smaller players at a competitive disadvantage. So, the logic is that bigger is better, and, as a corollary, smaller is weaker. For AT&T and Time Warner to get to that mid-2018 merger completion date will now require a legal victory in a U.S. District Court.

The official response from AT&T is this “(the) DOJ lawsuit is a radical and inexplicable departure from decades of antitrust precedent. Vertical mergers like this one are routinely approved because they benefit consumers without removing any competitor from the market. We see no legitimate reason for our merger to be treated differently” -  David R. McAtee II, Senior Executive Vice President and General Counsel, AT&T Inc. 

For network operators – bigger is better, especially with content

Since the time the proposed acquisition was announced in October 2016, AT&T has been arguing that the primary driver for the deal is to bring content and distribution under one roof. The merger will combine Time Warner's library of content and ability to create new premium content with AT&T's extensive customer relationships, world’s largest pay TV subscriber base and scale in TV, mobile and broadband distribution.

As a reminder, Time Warner, which was formed in 1990 through the merger of Time Inc. and Warner Communications, encompasses many premium media properties, including HBO, New Line Cinema, Turner Broadcasting System, The CW Television Network, Warner Bros., CNN, Cartoon Network, Boomerang, Adult Swim, DC Comics, Warner Bros. Animation, Castle Rock Entertainment, Cartoon Network Studios, Esporte Interativo, Hanna-Barbera Productions, Warner Bros. Interactive Entertainment. It also owns 10% of Hulu.

The basic idea driving the merger is for Time Warner to act as the content arm for AT&T, providing mobile and fixed broadband line subscribers with valuable material as part of a packaged service bundle. Consumers presumably would purchase an AT&T service bundle based on the perceived quality and value of the package rather than simply the lowest price for mobile connectivity. This will allow ARPU to rise and ensure a 'stickiness' factor that goes beyond the latest mobile handset deals, currently a leading cause for subscriber churn.

So, until we hear otherwise or until the courts rule that the merger is impermissible, the presumption is that “bigger is better” and that AT&T and Time Warner will continue to pursue their business combination.

A mobile + cable merger in Sweden

Earlier this week, another merger was proposed also on the premise that bigger is better. Tele2 and Com Hem agreed to a merger that will create the second largest mobile telephony and fixed broadband provider in Sweden (after Telia) and the market leader in digital TV. Com Hem’s shareholders will receive as merger consideration SEK 37.02 in cash plus 1.0374x new B shares in Tele2 for each share in Com Hem. This values the deal at about US$3.3 billion.

Com Hem operates a fiber-coax network serving approximately 1.5 million residential customers across Sweden. The company was established in 1983 and has approximately 1,200 employees. Its head office is in Stockholm.

Tele2, which was established in 1993 and is based in the Kista Science City, Stockholm, Sweden, operates an extensive mobile network across Sweden and has interests in The Netherlands, Lithuania, Latvia, Estonia, Kazakhstan, Croatia, and Germany.

The combined company will have a customer base of 3.9 million mobile customers, 0.8 million broadband customers, and 1.1 million digital TV customers in Sweden. Its 4G network will cover the entire country while its broadband network will cover almost 60 percent of Sweden’s households.
In presenting their merger to investors and to the press, Tele2 officials spoke of “evolving customer needs” and the appetite for digital content. As with the AT&T + Time Warner deal, there is an impetus to bring mobile, broadband and TV content under one roof.  

Some Service Providers are downsizing

One network operator moving in the opposite direction. Altice, the French operator led by business tycoon Patrick Drahi, who is known for ownership of his French cable operator Numericable.
Through a series of deals, in 2013 Drahi acquired SFR, France’s second largest mobile phone and internet provider from Vivendi. In late 2014, Altice acquired Virgin Mobile France for €325 million. The following year, Altice acquired Portugal Telecom and sold Cabovisão to Apax France. The hunger to grow bigger continued with a bid to acquire Bouygues Telecom, the third largest telecoms company in France. This merger was rejected by Bouygues Telecom. By then, Drahi had his sights on the U.S. cable market. In May 2015, Altice spent $9.1 billion to acquire a 70% controlling stake in Suddenlink Communications, which valued the seventh-largest U.S. cable company. This was soon followed in September 2015 with a $17.7 billion deal to acquire Cablevision, the dominant cable operator in the New York metropolitan area market. This deal was consummated in June 2016, making the new Altice USA into the #4 cable operator in the U.S. with more than 4.6 million Cablevision and Suddenlink customers across 20 states.

Many of the deals were accomplished with private equity debt. Now, 18 months after the transaction was completed, it appears that Altice has a case of indigestion. Perhaps bigger is not better, or maybe compelling content synergies have not been found across these diverse markets. Is there enough content synergy between France and New York to truly make one Altice brand?

This week, Altice N.V. announced a corporate restructuring centered on the separation of Altice USA from Altice Europe. The separation is to be effected by a spin-off of Altice NV’s 67.2% interest in Altice USA through a distribution in kind to Altice NV shareholders. Following the spinoff, the two companies will be led by separate management teams. Patrick Drahi, who will retain control of both companies, issued the following statement: “The separation will allow both Altice Europe and Altice USA to focus on their respective operations and execute against their strategies, deliver value for shareholders, and realize their full potential. Both operations will have the fundamental Altice Model at their heart through my close personal involvement as well as that of the historic founding team."

Tuesday, January 23, 2018

Silicon wars heat up in 2018 - Qualcomm pushes into automotive

by James E. Carroll

You might not expect the annual Consumer Electronics Show in Las Vegas to be a showcase opportunity for silicon players such as Intel, Qualcomm or NVIDIA. After all, Mobile World Congress 2018 opens in Barcelona in less than 50 days and it is here that we expect to see the latest cellular and Wi-Fi technologies. But the race is on to build new ecosystems for autonomous vehicles, smart cities, connected homes, etc.

In the automotive sector, Qualcomm holds a strategic position with its LTE modems which are currently used in millions of 2018 models from most of the major auto manufacturers. Through its Mobileye acquisition, Intel holds a strong position with next-gen sensors for autonomous vehicle functions. Meanwhile, NVIDIA established an early lead with SoC solutions for the digital dashboards and instrument panels of high-end autos, and this is leading to opportunities to become the silicon platform for the AI-powered cockpits of future autonomous vehicles. All of the silicon players are aiming for this goal.

Many are predicting that AI-powered autonomous vehicles will become “smartphones on wheels” or “rolling data centres.” Some forecasts put the amount of data generated by an autonomous vehicle at upwards of 4 TB per day, which is not hard to imagine if each vehicle is equipped with a dozen HD video cameras and multiple LIDARs, not to mention the data consumption needs of multiple passengers each playing with their own entertainment system.

Qualcomm is making  inroads with Ford and BYD

Qualcomm and Ford are collaborating on the development of advanced connectivity systems for vehicles using Cellular Vehicle-to-Everything (C-V2X) technology.

C-V2X is designed to allow vehicles to communicate directly with other vehicles, pedestrian devices, and roadside infrastructure, such as traffic signs and construction zones, without the involvement of a cellular network, or cellular network subscription.

C-V2X field validations are expected to begin in 1H 2018 in San Diego, along with additional trials in Detroit.

Qualcomm's first C-V2X chipset is expected to be commercially available the second half of 2018.

Qualcomm and Ford are also working on automotive telematics platforms with integrated Qualcomm Snapdragon LTE modems.

Qualcomm also announced a major deal with BYD ( "Build Your Dreams"), the Shenzhen-based manufacturer known for its high volume production of electric automobiles, buses, forklifts, rechargeable batteries, trucks, etc., as well as for having attracted a $230 million investment from Warren Buffet back in 2008. BYD’s stock price has been booming as of late, especially after the Chinese government announced plans to phase out fossil fuel cars in favour of EVs.

Over the next few design cycles the requirements for in-vehicle displays are expected to include the need for sufficient bandwidth to stream high-definition videos onto very high-resolution displays, while supporting audio and video streaming from multiple devices through Wi-Fi or Bluetooth. We already see Teslas, Audis, BMWs and many other cars with large digital displays. BYDs electric cars will have to compete.

Under the deal announced this week at CES, Qualcomm’s Snapdragon 820A Automotive platform will be used for integrated infotainment and digital cluster systems in electric vehicles starting in 2019. The infotainment and digital cluster systems will be integrated into BYD’s single electronic control unit (ECU).

The Snapdragon 820A Automotive platform consists of customized Qualcomm Kryo CPU, Qualcomm Hexagon 680 DSP with Hexagon Vector eXtensions (HVX) and Qualcomm Adreno GPUs. The platform also supports vehicle sensors, which will be key to computer vision and driver assistance systems. Qualcomm is pursuing this too with its Snapdragon Neural Processing Engine.

“As infotainment systems become more relevant in purchase decisions, it is important that automakers are armed with the industry’s most comprehensive and advanced solutions,” said Nakul Duggal, vice president of product management, Qualcomm Technologies.”

Qualcomm Mesh Network for Smart Homes

There is a lot of CES buzz around smart speaker systems for the home, headphones and earbuds with Google Assistant or Amazon Alexa on-board. Many are interesting products with the potential to keep millions -- and someday billions—of consumers firmly attached to the digital empires of the hyperscale players. At the networking layer, all of these gizmos will rely on robust home networks. Today, most home set-ups include some sort of broadband modem connected to a single WiFi hotspot. That’s not enough, say the vendors! Why not build an in-home mesh with a WiFi repeater in every room?

Qualcomm Technologies is currently working on mesh networking with a number of start-ups, including Cognitive Systems Corp., Origin Wireless and Lunera.

  • Cognitive Systems offers its “Aura” WiFi Motion technology available for any product based on the Qualcomm Mesh Networking Platform.  Aura’s WiFi Motion software enables advanced motion detection using RF motion algorithms and machine learning.
  • Origin Wireless will be using the Qualcomm Mesh Networking Platform to further expand their motion-detection, vital sign detection, well-being monitoring and home security software to bring new use cases to their suite of smart home solutions.
  • Lunera, an IoT infrastructure company, will be utilizing the Qualcomm Mesh Networking Platform to upgrade the connectivity in their newly announced Lunera Ambient Compute software platform, which can connect billions of IoT devices to the cloud. 

Monday, January 22, 2018

Silicon wars heat up in 2018 – Intel at CES

With the start of the new year comes the massive Consumer Electronics Show in Las Vegas and with it a flurry of technology announcements from the major silicon companies. The publicity focus is not just on shiny new electronics but on a range of new technologies driving cloud services, fog computing, edge data centres and the future of network connectivity. A big battle is clearly shaping up for brains of autonomous vehicles. The same can be said for the connected home, smart cities, healthcare, entertainment, gaming, etc.

Intel takes the stage

Before launching into his annual CES keynote on the near-term future of processing, Intel’s Brian Krzanich first needed to publicly respond to the “Spectre” and “Meltdown” security threats that have dominated tech news. He said Intel has not yet seen a real case where these vulnerabilities have led to a cyber exploit. Nevertheless, Intel will issue software patches for 90% of its products in the coming days with the remainder expecting fixes too. Intel is going to some lengths to assure customers that these software patches will have minimal performance impacts. Already, Amazon Web Services, Microsoft, and Google are stating that their public cloud computing resources are secure and experiencing only minimal performance variations. For assurance, Intel cites the following:

Apple“Our testing with public benchmarks has shown that the changes in the December 2017 updates resulted in no measurable reduction in the performance of macOS and iOS as measured by the GeekBench 4 benchmark, or in common Web browsing benchmarks such as Speedometer, JetStream, and ARES-6.”
Microsoft“The majority of Azure customers should not see a noticeable performance impact with this update. We’ve worked to optimize the CPU and disk I/O path and are not seeing noticeable performance impact after the fix has been applied.”
Amazon“We have not observed meaningful performance impact for the overwhelming majority of EC2 workloads.”
Google“On most of our workloads, including our cloud infrastructure, we see negligible impact on performance.”
This may not be the end of this trouble for Intel nor for its CEO, who is facing questions about his reported sale of stock options during the period before the vulnerabilities were published.

The data tsunami will benefit society

Krzanich used his CES keynote to highlight opportunities brought about by the explosion of data. It is the tsunami of data that is driving the next great wave of the technology revolution and with it, profound social change.  

“Data is going to introduce social and economic changes that we see perhaps once or twice in a century,” Krzanich said. “We not only find data everywhere today, but it will be the creative force behind the innovations of the future. Data is going to redefine how we experience life – in our work, in our homes, how we travel, and how we enjoy sports and entertainment.”

For technologists, the main questions are how much data, where does the processing occur, and what are the storage and networking requirements? For example, Intel says:

  • Average Internet users are consuming 1.5 GB per day
  • Autonomous vehicles will generate about 4 TB per day
  • A connected airplane will generate 40 TB per day
  • A smart factory could general 1 petabyte of data, the equivalent of data production equivalent of 700,000 people playing with smartphones.

Conventionally, data has been stored for later processing. Today’s applications increasing presume that it will be processed and analysed in real time.  Virtual reality and augmented reality consumption devices will need to perform real-time stitching between multiple video streams. Only some of the terabytes of data need to be carried over the network but the latency requirements will be tight.

Immersive Media

For CES, Intel is putting a heavy emphasis on “immersive media.” For example, one experiment being conducted with the National Football League places dozens of connected, 5K video cameras around an athletic field. The image streams are stitched together in real time. The system calculates volumetric pixels – called voxels – which can be viewed from any angle, creating an immersive experience for the viewer.  The first set-up in this experiment is producing 3 TB of data per minute.
Krzanich also announced the launch of Intel Studios, which has just completed the construction of a state-of-the-art sound/video stage in Los Angeles that is capable of stitching video streams from 100 cameras. The 10,000-square-foot dome is described as “the world’s largest volumetric video stage.” The post-production room is equipped with Intel-powered graphics workstations and servers with the ability to crunch over 1TB of data every 10 seconds. Paramount Pictures is the first major Hollywood studio to sign up as a partner.

Neuromorphic computing

Intel Labs has developed a prototype chip called “Loihi” that is based on the principles of neuromorphic computing, meaning that it aims to mimic the processing of the human brain.  Loihi combines training and inference on a single chip. The researchers say it mimics the natural learning process by forming new connections between neurons, or reprogramming the transistor connectivity on chip. The Loihi chip is currently capable of rudimentary image recognition in the lab. Intel plans to share it with research institutes later this year.

Intel is also announcing the design, fabrication and delivery of “Tangle Lake,” a 49-qubit superconducting quantum test chip. This is quite a gain over the 17-qubit design that Intel announced a few months back, but clearly still a prototype for ongoing research. Krzanich said Intel’s goal is to produce a complete quantum computing system – from architecture to algorithms to control electronics. The company reckons that we are still five to seven years away from addressing the manufacturing challenges that would have to be overcome for a commercial product. Getting to a commercial system will probably require chips with one million or more qubits on board. Delph University in the Netherlands is one of Intel’s research partners.

Moving fast with autonomous vehicles

The highlight of the 2018 CES show for Intel is its progress with autonomous vehicles. It was just about one year ago that Intel agreed to acquire Mobileye, a developer of machine vision systems for automated driving, for about $15 billion.

Mobileye, which is based in Israel and is now a wholly-owned division of Intel, holds the leading market position in computer vision for Advanced Driver Assistance Systems (ADAS). Its portfolio includes surround vision, sensor fusion, mapping, and driving policy products. Mobileye's EyeQ chips are already installed in about 20 million vehicles.

This installed base of Mobileye vehicles provides a strategic crowdsourcing mechanism for Intel and its auto manufacturing partners to develop the highly accurate maps needed to gain centimetre precision in the guidance of autonomous vehicles. Up to 2 million vehicles from BMW, Nissan and Volkswagen are now expected to use the Mobileye Road Experience Management (REM) technology to crowdsource this type of data.

Mobileye’s upcoming EyeQ4 and EyeQ5 chips for Level 3/4 autonomous driving programs go into production in 2018 and 2020 respectively. Mobileye currently has OEM relationships with GM, VW, Honda, BMW, PSA, Audi, Kia, Nissan, Volvo, Ford, Renault, Chrysler, SAIC and Hyundai.  Intel’s latest automated driving platform combines automotive-grade Intel Atom processors with Mobileye EyeQ5 chips to deliver a platform for L3 (Level 3) to L5 (Level 5) autonomous driving.

Flying taxis too

Intel is working with Volocopter, a start-up based in Germany that is developing autonomous, fully-electric, vertical take-off flying machines. The company plans to offer air taxi services in major cities.  The prototype uses the same technology that Intel is supplying to drone manufacturers.

Splitting ways with Micro on future 3D NAND

While most of Intel’s CES news is about building partnerships, there was one item moving in the opposite direction. Micron and Intel agreed to work independently on future generations of 3D NAND. The companies had previously been engaged in a partnership for NAND memory and are currently ramping products based on their second-generation of 3D NAND (64 layer) technology. The new business arrangement will go into effect after the companies complete development of their third-generation of 3D NAND technology, which will be delivered toward the end of this year and extending into early 2019.  Neither expects change in the cadence of their respective 3D NAND technology development of future nodes. Intel and Micron will also continue to jointly develop and manufacture 3D XPoint at their joint venture fab in Lehi, Utah, which is now entirely focused on 3D XPoint memory production.

Will Huawei's rapid growth continue in 2018?

by James E. Carroll

For the past decade, Huawei has been a shining star for the networking and telecoms field. 2015 and 2016 were especially good years as the company’s overall revenue grew 37% and 32%, respectively.
In a New Year’s message to staff, Huawei’s rotating CEO Ken Hu noted that the tepid growth rate slowed considerably, to 15% in 2017.  This is the slowest pace of expansion since 2013. Huawei's final revenue figure for 2017 should be in the range of 600 billion yuan (US$91 billion). Hu cited fluctuations in telco investment cycles, perhaps alluding the strange position that the Chinese telecom sector finds itself now that 4G infrastructure rollouts are mostly completed but 5G has yet to arrive.

Still, Huawei remains strong compared to any of its nearest competitors. For its most recently completed fiscal quarter, Cisco reported a growth rate of -2%.  Nokia reported a 7% year-on-year decrease (4% decrease on a constant currency basis).  Ericsson’s Q3 sales were also down 6% year-over-year (down 3% on a constant currency basis). Even when looking ahead to 2020, Ericsson is forecasting that its sales of radio access network equipment will decline or at best remain flat. At its Ericsson Capital Markets Day in Stockholm in November, the company even gloomily predicted that sales for IT and cloud solutions would likely decline.  Even worse, in late December, Ericsson signed a credit agreement with the Nordic Investment Bank (NIB) for US$220 million and another one with AB Svensk Exportkredit (SEK) for US$150 million, to shore up its balance sheet and help fund its 5G R&D activities.

These discouraging pictures come after several years of weak Service Provider sales for the industry as a whole. Ericsson's revenue crashed 10% YoY in 2016. For instance, IBM's sales in 2016 were down 2%, Microsoft's were down 9%, and Cisco (despite spending several billion dollars on expensive acquisitions every year) struggles to even maintain its current level of sales, although to be fair, profitability has been improving.

Despite the slide in its growth rate, Huawei’s CEO said the overall business remains strong. Unlike any of its network equipment rivals, Huawei also competes in the consumer segment, where the smartphone is king. In 2017, Huawei sold a record 153 million smartphones, giving it approximately 10% of the global market. The flagship Huawei Mate 10, which boasts an in-house 8-core CPU and 12-core GPU based on 10nm technology, is critically regarded as a close competitor to the flagship phones of Apple and Samsung. It is also priced at a significant discount to those brands. It now looks quite possible for Huawei to establish itself as a premier global brand. Remember when Nokia enjoyed such success? Ericsson too was a mobile phone contender in the days before the iPhone. Cisco even made a half-hearted effort to enter the consumer space. Now, all of these companies can only sit back and watch as Huawei uses its growing consumer success to build up its brand.
In the enterprise sector, Huawei is focused on opportunities in cloud, campus networks, data centres, and IoT. Hu said 197 companies in the Fortune Global 500 have selected Huawei as their digital transformation partner.

Hu’s New Year’s greeting did not provide a geographic breakdown of the company’s sales (perhaps we will get that information in a later report), but over the past few years, these have been fairly evenly split between domestic and international sales. In its home market, we can confidently say that Huawei’s position is strong and getting stronger. While the western vendors continue to compete in China via their joint venture companies, the nationalist tendency to “buy local instead of imports” seems to be on the rise.

One example is China Telecom recently completed 100G ROADM backbone network on the middle and lower reaches of the Yangtze River, including the provinces of Jiangsu, Zhejiang,  Hubei, Anhui, Jiangxi, and the municipality of Shanghai.  In the past, this type of contract might have split between several vendors, with perhaps a small share of the pie going to one of the foreign joint venture companies. That was not the outcome here. Huawei was the exclusive network solutions provider for the China Telecom optical backbone project. Remarkably, the completed the upgrade in only five months, which is perhaps indicates a good contracting decision by China Telecom. Since the project is described as an optical mesh backbone that is already carrying over three hundred 100G services from the outset, we can presume that the network will grow by leaps and bounds over the coming decade – the Yangtze River basin is after all one of the most densely populated regions of the planet.  Good luck to any other vendors getting in

In the international market, we only have anecdotal evidence for how Huawei is actually performing. Because Huawei is a privately-held company, only a limited amount of financial data is disclosed to the market. The biggest story in networking is, of course, the rapid rise of the public cloud, especially AWS, Microsoft, Google, Alibaba and handful of others. To our knowledge, Huawei is not a significant supplier to any of the hyperscalers.

Although Huawei has made remarkable inroads with top telecom operators in western Europe, its revenue growth rate is constrained by the flat to declining CAPEX budgets. Until investments in network infrastructure rise again, it will be difficult for any vendor to grow faster than 10 per cent annually.

There are opportunities for Huawei in developing markets, especially those tied into China’s Belt and Road initiative.  For example, Huawei Marine is the lead vendor for a new Pakistan East Africa Cable Express (PEACE) submarine cable that will connect South Asia with East Africa. This project will  offer the shortest fibre route from western China to southern Europe, when combined with terrestrial fibre between Pakistan and China. The project is funded by China Construction Bank.

One problem that has not been solved is the impasse with the U.S. government. Over five years have now passed since the U.S. House of Representatives' Permanent Select Committee on Intelligence issued its report recommending that Huawei and ZTE be blocked as suppliers of critical infrastructure due to security concerns. The United States remains the largest single market. No resolution to this standoff appears to be in sight. More recently, the Australian intelligence service has acted to block Huawei Marine from its role in funding and constructing a high-capacity subsea cable that is to link Sydney to the Solomon Islands. Apparently, security concerns were raised concerning China’s growing presence in the region. Nevertheless, Huawei smartphones are growing in popularity with consumers worldwide, including in the U.S. and Australia.

Monday, January 8, 2018

Will Europe regain mobile leadership with 5G?

by James E. Carroll

At the end of November, Verizon confirmed plans to launch 5G residential service beginning in Sacramento, California. By the end of 2018, Verizon should have commercial 5G fixed line residential services running in half a dozen cities. The timing is a little bit slower than what was suggested at Mobile World Congress this year, but it is is a very aggressive rollout plan. AT&T has been testing fixed line 5G services in several markets too, and we can expect a similar early entrance into the 5G commercial arena. So, 5G will be coming to U.S. markets over the next 18 months with a credible use case and business plan.

Even before these two operators stake their claims of being the 5G deployment leader, we should expect the Korean operators to showcase 5G small cell technology at the PyeongChang 2018 Winter Olympics, although this event – assuming it is not disrupted­­­ by the threat of war – perhaps will be more of a public demonstration rather than a commercial deployment. ­There are, of course, many publicity events for 5G occurring around the globe. As previously reported in this journal, NTT DOCOMO has various 5G experiments underway in its home market, including 5G for autonomous vehicles. This application could turn out to be one of the best use cases for 5G given the need for extremely low latency when navigating a car in city traffic. So, Korea and Japan are the likely first movers with 5G.

The question has been raised by many: will Europe regain its leadership in mobile technologies with 5G? This topic was the subject of a meeting of EU Telecom ministers held last week in Tallinn, Estonia. The outcome of the meeting was the adoption of a 5G action plan -- or roadmap – aimed at spurring widespread deployment of 5G in EU nations by 2025. Naturally, there is no budget assigned to make it happen – that will be up to Service Providers to incorporate into their CAPEX plans. The roadmap is simply the next steps that the EU plans to take to harmonise spectrum bands for 5G.

"The 5G roadmap lays out major activities and their time frame. With the roadmap we agreed on plans for harmonising the technical use and purpose of 5G spectrum and their allocation to telecommunications operators. It is no secret that Digital Europe is a priority for the Estonian presidency. However, a digital society cannot be created without 5G networks," commented minister Urve Palo, Minister for Entrepreneurship and Information Technology for Estonia. "By the year 2025, we want to see the presence of 5G connectivity in large cities and along major transport routes of every European country. 5G networks are needed both for citizens and the devices that need reliable and high-speed Internet access to cope with increasingly large quantities of data.”

Some catching up will be required

A new report from 5G Americas, which is a trade association, finds that North America continues to hold a lead over other regions in terms of overall LTE adoption to the greatest percentage of users. Citing data from Ovum, 5G Americas reports that the number of active LTE connections worldwide passed 2.5 billion as of the end of the third quarter of 2017. This is out of a total of 7.8 billion total cellular connections worldwide. The report states that North America achieved 341 million LTE subscriptions by the end of September 2017 “with some of the highest penetration rates, most extensive coverage and largest market share for LTE in the world.” 

The LTE penetration rate for North America was 94 percent in the third quarter, with 341 million connections compared to the population of 362 million in North America. In comparison, Western Europe was found to have a 57 percent penetration for LTE. Given that LTE commercial service rollouts began in 2011, it is somewhat puzzling that nearly seven years later, such a significant percent of the population in Western Europe either hasn’t gotten around to updating to a 4G-enabled phone, or simply hasn’t been interested in subscribing to an LTE plan. The reason most often stated for this slow adoption has been cost. Data plans in Europe are simply costlier and more restrictive that in other developed markets. As we’ve seen recently in India with the Reliance JIO 4G launch, if unlimited data is offered at the right price, consumers will respond in droves.

Highlights of the 5G Americas report (data from Ovum)

  • There was an increase of 838 million new LTE subscriptions in one year ending September 2017.

  • Global market share for LTE achieved 32 percent at the end of the third quarter of 2017, an increase of almost 10 percentage points in one year.

  • As of mid-November, there were 562 commercial LTE deployments worldwide, while 211 of those operators have already evolved to LTE-Advanced (TeleGeography).

  • In Oceania, Eastern and Southeastern Asia, LTE penetration is at 58 percent

  • In Western Europe, LTE penetration is at 57 percent penetration.

  • Latin America nearly doubled LTE connections to 179 million from 99 million year-over-year at 3Q 2017 increasing by 80 percent.  LTE’s market share increased from 14 percent to 26 percent in twelve months out of a total mobile subscription base of 691 million.

  • As of mid-November, there were 108 commercial LTE networks across the Latin America and the Caribbean region of which 20 are LTE-Advanced deployments with Carrier Aggregation (TeleGeography).