Using Bigger Bandwidth Applications

The recent Cisco Annual Internet Report for 2018 – 2023 had one chart that I found intriguing. The purpose of Cisco’s report is to look at the future of broadband usage, and the report included a chart showing the amount of bandwidth needed for various web functions. To me this list was reminiscent of the list that the FCC made in 2015 when they set the definition of broadband at 25/3 Mbps – except that all of the item on this list require more bandwidth than the functions the FCC foresaw just five years ago.

I think Cisco’s point is that we find ways to use more broadband when it’s available. Most of the items on this list are already in use today, with the last few on this expected in the near future.

  • 4K security cameras – 16 Mbps
  • Streaming 4K Video – 16 Mbps
  • Virtual Reality Streaming – 17 Mbps
  • Self-Driving Car Diagnostics (done when your car pulls into the driveway) – 20 Mbps
  • Cloud Gaming – 30 Mbps
  • Streaming 8K Video – 51 Mbps
  • 8K Wall TV – 100 Mbps
  • Online HD Virtual Reality – 167 Mbps
  • Online 8K Virtual Reality – 500 Mbps

Cisco notes that these functions have become viable for the public as the amount of bandwidth to homes has grown. Anybody with broadband speeds of 125 Mbps or faster should be able to use all except the last few services. In the US a lot of homes now have Internet speeds in this range as Comcast, Charter and the other big cable companies have increased basic speeds to 100 – 200 Mbps homes with the introduction of DOCSIS 3.1. Charter increased my home last year from 60 Mbps to 135 Mbps.

4K security cameras have been on the market for a few years. They provide enough resolution to clearly identify people at the front door or outside a factory. The 16 Mbps bandwidth application is needed to upload video images into the cloud or to view the camera feed remotely over the web. Interestingly, a 4K security camera ought to have a fast upload speed to work properly – something that is still lacking for most cable company connections.

The web is now full of 4K videos on Netflix, Amazon Prime, YouTube and elsewhere. There are already web sites doing virtual reality streaming.

The self-driving car diagnostics is an interesting category. My wife’s 2019 Subaru already connects to the web every time she pulls into the driveway. This connection is likely not at 20 Mbps, but her car is doing diagnostics and uploading the results of driving to Subaru, and also making driving statistics available to us.

Cloud gaming is already here, although most applications are streaming at 4K or slower speeds. However, since several of the game companies have migrated online, the intensity and bandwidth needed for games is expected to increase, and Cisco pegs that at needed a 30 Mbps connection. What this speed requirement doesn’t tell you is that kids that routinely run several games simultaneously.

Bandwidth needs really shoot up for 8k video at 51 Mbps per stream. 8K content is already widely available on YouTube and other websites. 8K TVs have been around for a few years. Their prediction of 100 Mbps for an 8K TV assumes multiple simultaneous streams – something that sports fans routinely do.

Cisco also lists two near-future applications that will be real bandwidth hogs. They estimate that HD virtual reality done online will require 167 Mbps. They set 8K virtual reality as needing 500 Mbps. These functions are going to need faster broadband connections than what most homes have today. However, OpenVault reports that the number of US homes subscribing to a gigabit connection doubled in 2019 to 2.8% of all households. As that number keeps growing there will finally be a market for applications that need giant bandwidth. For years the industry has searched for gigabit applications, but nobody developed them since there have been so few homes that could use them. 8K virtual reality would bring true immersive virtual reality into the home – but ISPs are going to hate it. They love selling gigabit connections, but they don’t really expect homes to use that much bandwidth.

The Explosive Growth of M2M Traffic

The Cisco Annual Internet Report for 2018 – 2023 is full of interesting predictions this year. One of the more intriguing predictions is that Machine-to-Machine (M2M) traffic (which they also refer to as Internet of Things (IoT) traffic) will become a little more than half all of the traffic on the web by 2023. That’s an amazing prediction until you stop and think about all of the devices that communicate with the Internet without needing a human interface.

Cisco forecasts several reasons why M2M traffic will grow so much in the next few years. The primary way is through the proliferation of M2M devices. They predict over the 5-year period there will be a 2.4 times growth in connected devices from 6.1 billion in 2018 to 14.7 billion in 2023. That’s a 19% compounded growth rate and by 2023 equals 1.8 connected devices for every person on earth.

The second reason for the growth is that we are using M2M devices for a lot more functions than just a few years ago. Cisco is predicting fast growth in the following categories of M2M

  • They predict the number of worldwide connected home devices will grow by 20% per year. This is the largest category of devices and will represent just under 50% of connected devices by 2023. This category includes things like home automation, home security and video surveillance, connected white goods (the new term for connected appliances), and our communications and entertainment devices like smart TVs, laptops, desktops, and smartphones.
  • They predict that connected car applications will be the fastest-growing sector, growing at 30% per year. This includes connections made for things like fleet management, in-vehicle entertainment, emergency calling systems, vehicle diagnostics, and vehicle navigation.
  • Cisco predicted that connected city applications will be the second fastest-growing M2M category with a 26% compounded growth. This includes things like smart traffic systems, surveillance cameras, weather and environmental monitors, smart parking systems, gunshot monitors, etc.
  • They predict that connected health will grow 19% annually. This category mostly consists of telemedicine monitors used for outpatient monitoring.
  • Connected energy applications are predicted to grow by 24%. This includes smart grid monitors that track utility usage and loads, and pinpoint network outages quickly. It includes energy monitors, which can turn off air conditioners during times of heavy peak usage. In includes sensors in water systems that track pressure and usage and that predict underground leak locations.
  • Cisco predicts connected work will grow by 15%. This is used for things like inventory tracking, surveillance and security monitoring, and tracking and connecting to employees working in the field.
  • They predict that connected retail will grow by 11% annually. M2M traffic is being used to track inventory. Big chain stores are starting to track the shopping pattern of individual shoppers to see how they traverse the various departments.
  • Connected Manufacturing and Supply Chain will grow by 8% annually. Supply chain monitoring tracks the status of delivery for components needed in the manufacturing process. This also includes smart warehousing that automates packing and shipping or orders. Smart manufacturing supports monitors that track the performance of machinery and manufacturing processes.
  • They predict all other M2M traffic will grow by 19%. This would include things like smart agriculture where monitors are tracking individual herd animals and are just starting to be deployed to monitor crop conditions. This would include other things like sports monitors.

The volume of traffic generated by M2M traffic surprises people. So much of what we do happens in the background and we either forget about it or don’t even know it’s happening. For example, there was an article in the Washington Post last year by a reporter that left the country for a month and left his cellphone in his home. During his absence, the phone used a significant portion of his monthly data plan by updating apps and communicating regularly with remote web sites. My wife’s car connects to the web through or WiFi every time she pulls into the driveway and uploads diagnostics of the various monitors and checks for and downloads needed software updates. Whether for good or bad, our machines and electronics are connecting to the web and using broadband.

Broadband is Now a Mature Market

One of the most interesting things revealed by Cisco’s latest Annual Internet Report is the extent to which North America is now largely a mature broadband market. In this case, North America is the combination of the US and Canada and Cisco does not provide the data for each country.

Consider the following statistics for North America between 2018 and 2023:

  • The percentage of people using the Internet in 2023 will grow to 92% (345 million users) up from 90% (328 million users) in 2018.
  • The percentage of people using cellphones in 2023 will grow to 88% (329 million users) up from 86% (313 million users) in 2018.

This paints a picture of North America as a mature broadband market. While there are still new customers to land, ISPs collectively will not be winning many new customers. The growth of landline users from 328 million to 345 million over 5 years represents an annual growth rate of only 1%. In economic terms that’s a mature market.

There will likely continue to be movement within the market. In the year ending in the third quarter of 2019, the big cable companies took at least 2 million customers from telcos – a trend that is likely to continue. However, some telcos are fighting back by building fiber, such as the 12 million fiber passings built by AT&T over the last few years.

Cisco is painting the same bleak future for cellular customers and is predicting the same slow 1% annual growth for North America. The cellular companies have been waging a marketing war and stealing customers from each other – a largely zero-sum game. The cellular market is getting tougher as Comcast and Charter continue to win cellular customers and Dish Networks is poised to enter the market in a few years.

The mature nature of the broadband and cellular industries exposes the FCC’s fiction that carriers will be spending a lot more capital due to relaxed regulations. It’s hard for any ISP to justify spending a lot of capital in a stagnant and slow-growing market. Any capital spending is being done to upgrade to newer technologies, but there is not a lot of capital needed due to customer growth.

Equipment manufacturers aren’t focusing on North America. While the US will add a net of 17 million people to the Internet over 5 years, Asia Pacific will be adding a billion new people, the Middle East and Africa will be adding 230 million people, and Latin America will be adding 83 million people. Europe is also a mature market and will only be adding 25 million people to the Internet over 5 years.

These numbers show why the administrations attempt to somehow squelch Huawei is likely doomed to failure. Huawei doesn’t need North America or Europe to succeed and can far outstrip European and American vendors by concentrating on Asia and Africa.

This slow growth also highlights the dilemma of the publicly traded ISPs. With a 1% annual growth rate the big ISPs start to look like electric utilities in terms of growth potential. Comcast and Charter  are still meeting Wall Street expectations due to taking customers from DSL, but even that growth has to slow and shrink away at some point. All of the other big cable companies are faced with trying to please Wall Street with stagnant customer counts – something that only can be done by raising rates, cutting costs through mergers, or introducing new revenue streams.

The industry doesn’t have far to grow after 2023. Numerous surveys have shown that most of the people that don’t buy Internet access either can’t afford it or live in a rural market where it’s not available. Since the big ISPs aren’t chasing either of those customer segments they are already collectively at their peak. Growth now comes only from general population growth – and even that news is not great as the US birth rate keeps dropping and immigration has been curtailed.

There is nothing wrong with a mature market from an economics perspective. Unfortunately, the high growth of broadband customers over the past twenty years has created an expectation on Wall Street that telecom companies have fast growth potential. I look at the basic numbers and wonder how long it will be until Wall Street resets that expectation. We’d all be a lot better off if the big ISPs didn’t feel huge pressure to grow the bottom line.

The Growth Rate of Broadband Speeds

Cisco has changed the name of its periodic predictions of broadband usage from the Visual Networking Index to the Annual Broadband Report, and recently issued a report that covers the period from 2018 to predictions made through 2023.

Cisco is one of the few industry players that projects future broadband usage. Their past reports have been spot on in terms of predicting future broadband usage.

One of the items forecast in the Cisco report this year is average landline Internet speeds by world region. Following is their prediction of the average broadband speeds (in Mbps) for North America. This represents a 20% compounded growth, just a hair slower than the 21% predicted in their 2019 report.

2018 2019 2020 2021 2022 2023
56.6 70.1 92.7 106.8 126.0 141.8

It’s worth noting that Cisco includes Canada along with the US in defining North America. I haven’t found equivalent numbers for Canada alone to know if they pull the composite number upward or downward. The big takeaway from the Cisco numbers is that broadband speeds are continuing to climb as ISPs either arbitrarily increase speeds or customers upgrade to faster networks.

Cisco also predicts the future of cellular broadband speeds, as follows:

2018 2019 2020 2021 2022 2023
23.6 31.2 40.1 48.2 54.4 62.4

Interestingly, the cellular speeds are faster than what has been reported by Opensignal. They reported average cellular speeds in early 2019 for the US carriers as AT&T – 17.8 Mbps, Verizon – 20.9 Mbps, T-Mobile – 21.1 Mbps, and Sprint – 13.9 Mbps. That’s slower than Cisco’s 2018  speeds – but it’s worth noting that Canada has one of the fastest cellular networks in the world, which probably raises the Cisco numbers. The bottom-line takeaway from the Cisco numbers is that cellular broadband speeds are growing at an average rate of 21% per year.

The Cisco numbers show that Cisco doesn’t buy into the story that 5G is going to massively increase cellular broadband speeds in the next few years. The most recent increases in broadband speeds come from a few factors. The big carriers are upgrading a lot of cell sites to full 4G and finally utilizing the full power of the 4G specifications. The recent proliferation of small cell sites is relieving congestion from tall cell sites, which should result in faster speeds. Some of the future speed increases are likely due to the phase-out of 3G. While 5G is obviously a component of future cellular speed increases we’re not likely to see a one-time spike in faster speeds.

Perhaps the biggest takeaways from the Cisco numbers is that the FCC is out of step with reality as they cling to the 25/3 Mbps definition of broadband. Cisco says the average landline broadband speed for all of North America in 2019 was 70 Mbps, climbing to almost 93 Mbps this year. Cisco estimates the average speed in three years at almost 143 Mbps. It’s hard to think of any possible justification for not increasing the definition of broadband to match the market.

Unfortunately there is one regulatory reason why the FCC won’t act. If they increase the definition of broadband, they will be declaring that millions of additional homes don’t have acceptable broadband. This particular FCC is not brave enough to take a black eye over the resulting headlines. They already fear that fixing their faulty broadband maps is going to uncover millions of additional rural households without adequate broadband.

The Resurgence of Wireless Mesh?

I’ve had several calls recently from clients asking about wireless mesh networks. Those that have been in the industry for a while probably remember the mesh network craze in the late 1990s. At that time large cities all over the country considered building WiFi mesh networks to try to bring broadband to everybody in their cities. Many cities deployed pilot systems, but in the end, the technology never panned out. The technology had the squirrely problems often associated with wireless technology and never delivered the bandwidth that the manufacturers promised.

Apparently, the technology is back. I went to the web for a quick investigation, and sure enough there are carrier-class outdoor mesh radios available from a number of different manufacturers. In case you aren’t familiar with the concept of a mesh network, it’s a network comprised of multiple radios, each of which connects to multiple other radios. Most mesh networks are dynamically linked, meaning that the radios work autonomously to find the most efficient routing path for traffic within the mesh. The easiest way to understand this is with this diagram from Cisco, which has been manufacturing mesh network gear for many years. In this diagram each radio interconnects with neighboring radios.

The biggest flaw in the technology two decades ago was that the mesh networks didn’t scale well. This was for two reasons. First, by definition, a wireless link loses half of its bandwidth with every hop to another radio. Mesh networks with too many hops don’t deliver very much bandwidth to the most remote nodes in the network.

Large mesh network also developed an unexpected problem. One of the characteristics of a mesh network is that the radios constantly coordinate with each other. If a given node is temporarily overloaded with a big bandwidth demand from an end user, the network dynamically routes other traffic around the bottleneck. Unfortunately, it turned out that in large networks the radios spent a majority of the bandwidth communicating with each other at the expense of the bandwidth left for end users. As mesh network grew in size the amount of bandwidth throughput decreased significantly. Technicians determined that this excess internode chatter could be reduced by limiting the number of nodes that any radio could communicate with, but in doing so the network was no longer a real mesh.

The other big problem in the 1990s is that the networks were deployed as outdoor radios, meaning that very little bandwidth actually made it into homes. I remember working one day at a client where I could see a nearby mesh radio through a window. As long as I sat where I had a direct line of sight to the radio I could use the WiFi, but if I moved to another part of the room the signal completely died. Broadcasting WiFi with outside radios is an inefficient way to provide bandwidth indoors.

Those inherent problems are still an issue today. There is no way to eliminate the issue of the bandwidth decreasing with each hop. However, the difference from today and the 19990s is that we can feed a mesh network with gigabits of broadband instead of with a few T1s. To some degree, that means that we can overpower the system so that at least some bandwidth makes it to the furthest nodes in the network.

One of the other weaknesses of a mesh network is that most networks use WiFi spectrum. Practically every wired home uses WiFi today to move bandwidth around the house. Superimposing a mesh WiFi network in a neighborhood means a lot more high-power WiFi sources to cause interference with every other WiFi device. Anybody who has ever tried to maintain a WiFi signal in a crowded business hotel understands the issues with WiFi interference.

Even with those limitations, I can see some great uses for a mesh network. The vendors are pushing the technology as a way to bring bandwidth more easily to outdoor spaces like parks. There is a brand of outdoor mesh devices being marketed as a way to spread WiFi around a farmhouse to the outdoor buildings. While nobody seems to be marketing the idea yet, a mesh network might be a good way to spread WiFi signals to fields and pastures to track the small bandwidth sensors being used to collect data from fields and herds.

What my clients really wanted to know is if a mesh network could be used to provide residential broadband. There might be situations where this makes sense. Rather than trying to beam the bandwidth from outside hotspots, each radio could feed a wire into a home. But mesh networks still have the same inherent problems as in the past and in most cases other solutions can probably produce faster and more consistent bandwidth. As a consultant I always have an open mind, but having seen the technology crash and burn once before I’d want to see this in practice before buying the resurgence of the technology again.

How We Use More Bandwidth

We’ve known for decades that the demand for broadband growth has been doubling every three years since 1980. Like at any time along that growth curve, there are those that look at the statistics and think that we are nearing the end of the growth curve. It’s hard for a lot of people to accept that bandwidth demand keeps growing on that steep curve.

But the growth is continuing. The company OpenVault measures broadband usage for big ISPs and they recently reported that the average monthly data use for households grew from 201.6 gigabytes in 2017 to 268.7 gigabytes in 2018 – a growth rate of 33%. What is astounding is the magnitude of growth, with an increase of 67.1 gigabytes in just a year. You don’t have to go back very many years to find a time when that number couldn’t have been imagined.

That kind of growth means that households are finding applications that use more bandwidth. Just in the last few weeks I saw several announcements that highlight how bandwidth consumptions keep growing. I wrote a blog last week describing how Google and Microsoft are migrating gaming to the cloud. Interactive gaming already uses a significant amount of bandwidth, but that usage is going to explode upwards when the machine operating the game is in a data center rather than on a local computer or game console. Google says most of its games will operate using 4K video, meaning a download speed of at least 25 Mbps for one stream plus an hourly download usage of 7.2 GB.

I also saw an announcement from Apple that the users of the Apple TV stick or box can now use it on Playstation Vue to watch up to four separate video steams simultaneously. That’s intended for the serious sports fan and there are plenty of households that would love to keep track of four sporting events at the same time. If the four separate video streams are broadcast in HD that would mean downloading 12 GB per hour. If the broadcasts are in 4K that would be an astounding 29 GB per hour.

The announcement that really caught my eye is that Samsung is now selling an 8K video-capable TV. It takes a screen of over 80 inches for the human eye to perceive any benefit from 8K video. There are no immediate plans for anybody to broadcast in 8K, but the same was true when the first 4K TVs were sold. When people buy these TVs, somebody is going to film and stream content in the format. I’m sure that 8K video will have some improved compression techniques, but without a new compression scheme, an 8K video stream is 16 times larger than an HD stream – meaning a theoretical download of 48 GB per hour.

Even without these new gadgets and toys, video usage is certainly the primary driver of the growth of household broadband. In 2014 only 1% of homes had a 4K-capable TV – the industry projects that to go over 50% by the end of this year. As recently as two years ago you had to search to find 4K programming. Today almost all original programming from Netflix, Amazon, and others is shot in 4K, and the web services automatically feed 4K speeds to any customer connection able to accept it. User-generated 4K video, often non-compressed, is all over YouTube. There are now 4K security cameras on the market, just when HD cameras have completely replaced older analog cameras.

Broadband usage is growing in other ways. Cisco projects machine-to-machine connections will represent 51% of all online connections by 2022, up from 40% today. Parks and Associates just reported that the average broadband home now has ten connected devices, and those devices all make internet connections on their own. Our computers and cellphone automatically update software over our broadband connections. Many of us set our devices to automatically back-up our hard drives, pictures, and videos in the cloud. Smart home devices constantly report back to the alarm monitoring service. None of these connections sound large, but in aggregate they really add up.

And sadly, we’re also growing more inefficient. As households download multiple streams of music, video, and file downloads we overload our WiFi connection and/or our broadband connection and thus request significant retransmission of missing or incomplete packets. I’ve seen estimates that this overhead can easily average 20% of the bandwidth used when households try to do multiple things at the same time.

I also know that when we look up a few years from now to see that broadband usage is still growing that there will be a new list of reasons for the growth. It may seem obvious, but when handed enough bandwidth, households are finding a way to use it.

Predicting 5G CAPEX

If you ever want a bad headache, spend a few hours researching predictions about the future trajectory of capital spending by the big players in the telecom industry. It’s a topic worth following since the big ISPs all said that eliminating net neutrality and other regulation would unleash them to spend lavishly on new networks.

I was looking through projections over the past year that were forecasting capital spending for 2019. My main motivation in looking at these projections was to see if the big companies are actually planning on spending money yet on 5G. I figured the best way to get past all of the 5G hype is to follow the advice from the movie All the President’s Men, and “Follow the money”.

I started by looking at projections from the beginning of 2018. The headlines at that time centered around the big benefits to the industry from the Tax Cut and Jobs Act passed in December 2017. That legislation created an annual benefit to AT&T of $2.2 billion and a benefit to Verizon of $4 billion annually. At the beginning of 2018 industry analysts predicted the companies would roll those savings into increased capital spending. However, like with most big corporations those savings were not rolled back into the business.

There were rosy predictions at the start of last year about 2018 and 2019 capital spending. For example, the analysts at Deutsche Bank Research said in February 2018 that capital spending by the wireless carriers would increase by 14% in 2018 and even more into the future. It’s not hard to understand the enthusiasm of the analysts because the carriers were fueling this story. Early in 2018 Verizon said they would be investing $35 billion in 5G and AT&T said they would invest $40 billion.

This enthusiasm was fueled all last year by promises from AT&T and Verizon to roll out 5G by the end of 2018. In June the analysts at Oppenheimer raised forecasts of capital spending for 2019 by $18 billion by Verizon and $25 billion for AT&T. However, at the end of last year we saw the 5G announcements had been nothing but hype when AT&T announced an imaginary 5G product and Verizon installed fixed wireless in a few hundred homes.

As recently as the fourth quarter of last year a number of analysts were still predicting greater capital spending for wireless this year compared to 2018. For example, MoffettNathanson LLC predicted 2019 capital spending would be up 3.3% in 2019. Most other analysts made similar projections.

As the books closed for 2018 it became obvious that the big wireless companies hadn’t spent nearly as much as expected for the year. For example, Verizon actual spending was $1.5 billion less than their own initial projections. AT&T came in $3 billion less than projected. When real spending materializes you start to understand the complexity of these budgets. For example, AT&T said that part of the reason for lower capital spending was due to delays in the deployment of FirstNet, the nationwide public safety network. I sit here wondering why FirstNet was even included in their capital budget since it’s not being funded from AT&T’s own revenues, but 100% by taxpayers.

By the time I got to looking at 2019 the picture gets incredibly muddled. There are still those predicting 2% to 3% more capital spending for 2019. But we also see the big carriers admitting to their investors that there will be little spending on 5G this year. This first big capital expenditure in 5G will be for the core electronics for 5G cell sites called the RAN. It doesn’t look like there will be a 5G RAN available for a few more years. Both cellular carriers admit that they are not spending much on 4G LTE infrastructure other than working on cell site densification in urban areas through the deployment of small cell sites aimed at relieving pressure on the big tower cell sites.

I’ll be honest that all I got out of this reading was the headache because I still have no idea about how much money these big carriers will spend this year. This is probably not abnormal in an industry under so much flux and I would imagine there are still decisions being made inside these companies every day that will change capital spending even in this year. The one thing I came away with was a clear picture that there will be very little spending in 2019 on 5G, which means that the announcements of the carriers to have 5G cellular products by 2020 are clearly still hype.

But there are still those in the industry with rosy predictions. The research firm IDC predicts that spending on 5G core equipment will increase worldwide from $500 million this year to $26 billion per year in 2022. Ericsson is predicting that 5G will account for 50% of mobile subscriptions in the US by 2023 along with a worldwide penetration at 20% that year. That seems to be in conflict with Cisco which recently predicted worldwide 5G penetration of 3% by the end of 2022. I have no idea which of these predictions is right, but I now know that we can’t put any faith in predictions about 5G spending or deployment, so perhaps all of this reading was not in vain.

The Continued Growth of Data Traffic

Every one of my clients continues to see explosive growth of data traffic on their broadband networks. For several years I’ve been citing a statistic used for many years by Cisco that says that household use of data has doubled every three years since 1980. In Cisco’s last Visual Networking Index published in 2017 the company predicted a slight slowdown in data growth to now double about every 3.5 years.

I searched the web for other predictions of data growth and found a report published by Seagate, also in 2017, titled Data Age 2025: The Evolution of Data to Life-Critical. This report was authored for Seagate by the consulting firm IDC.

The IDC report predicts that annual worldwide web data will grow from the 16 zettabytes of data used in 2016 to 163 zettabytes in 2025 – a tenfold increase in nine years. A zettabyte is a mind-numbingly large number that equals a trillion gigabytes. That increase means an annual compounded growth rate of 29.5%, which more than doubles web traffic every three years.

The most recent burst of overall data growth has come from the migration of video online. IDC expects online video to keep growing rapidly, but also foresees a number of other web uses that are going to increase data traffic by 2025. These include:

  • The continued evolution of data from business background to “life-critical”. IDC predicts that as much as 20% of all future data will become life-critical, meaning it will directly impact our daily lives, with nearly half of that data being hypercritical. As an example, they mention the example of how a computer crash today might cause us to lose a spreadsheet, but that data used to communicate with a self-driving car must be delivered accurately. They believe that the software needed to ensure such accuracy will vastly increase the volume of traffic on the web.
  • The proliferation of embedded systems and the IoT. Today most IoT devices generate tiny amounts of data. The big growth in IoT data will not come directly from the IoT devices and sensors in the world, but from the background systems that interpret this data and make it instantly usable.
  • The increasing use of mobile and real-time data. Again, using the self-driving car as an example, IDC predicts that more than 25% of data will be required in real-time, and the systems necessary to deliver real-time data will explode usage on networks.
  • Data usage from cognitive computing and artificial intelligence systems. IDC predicts that data generated by cognitive systems – machine learning, natural language processing and artificial intelligence – will generate more than 5 zettabytes by 2025.
  • Security systems. As we have more critical data being transmitted, the security systems needed to protect the data will generate big volumes of additional web traffic.

Interestingly, this predicted growth all comes from machine-to-machine communications that are a result of us moving more daily functions onto the web. Computers will be working in the background exchanging and interpreting data to support activities such as traveling in a self-driving car or chatting with somebody in another country using a real-time interpreter. We are already seeing the beginning stages of numerous technologies that will require big real time data.

Data growth of this magnitude is going to require our data networks to grow in capacity. I don’t know of any client network that is ready to handle a ten-fold increase in data traffic, and carriers will have to beef up backbone networks significantly over time. I have often seen clients invest in new backbone electronics that they hoped to be good for a decade, only to find the upgraded networks swamped within only a few years. It’s hard for network engineers and CEOs to fully grasp the impact of continued rapid data growth on our networks and it’s more common than not to underestimate future traffic growth.

This kind of data growth will also increase the pressure for faster end-user data speeds and more robust last-mile networks. If a rural 10 Mbps DSL line feels slow today, imagine how slow that will feel when urban connections are far faster than today. If the trends IDC foresees hold true, by 2025 there will be many homes needing and using gigabit connections. It’s common, even in the industry to scoff at the usefulness of residential gigabit connections, but when our use of data needs keeps doubling it’s inevitable that we will need gigabit speeds and beyond.

Fiber Electronics and International Politics

In February six us Intelligence agencies warned Americans against using cellphones made by Huawei, a Chinese manufacturer. They warned that the company is “beholden” to the Chinese government and that we shouldn’t trust their electronics.

Recently Sen Liz Cheney introduced a bill into Congress that would prohibit the US Government or any contractors working for it to use electronics from Huawei or from another Chinese company ZTE Corp. Additionally, any US military base would be prohibited from using any telecom provider who has equipment from these two vendors anywhere in their network.

For anybody who doesn’t know these two companies, they manufacture a wide array of telecom gear. ZTE is one of the five largest cellphone makers in the world. They also make electronics for cellular networks, FTTP networks and long-haul fiber electronics. The company sells under it’s own name, but also OEMs equipment for a number of other vendors. That might make it hard for a carrier to know if they have gear originally manufactured by the company.

Huawei is even larger and is the largest maker of telecom electronics in the world, having passed Ericsson a decade ago. The company’s founder has close ties to the Chinese government and their electronics have been used to build much of the huge wireless and FTTP networks in China. The company makes cellphones, FTTP equipment and also is an innovator in equipment that can be used to upgrade cable HFC network.

This is not the first time that there has been questions about the security of electronics. In 2014 Edward Snowden released documents that showed that the NSA had been planting backdoor software into Cisco routers being exported overseas from the US and that these backdoors could be used to monitor internet usage and emails passing through the routers. Cisco says that they had no idea that this practice was occurring and that it was being added to their equipment after it left their control.

Huawei and ZTE Corp also say that they are not monitoring users of their equipment. I would assume that the NSA and FBI have some evidence that at least the cellphones from these companies can be used to somehow monitor customers.

It must be hard to be a telecom company somewhere outside of the US and China because our two countries make much of the telecom gear in wide use. I have to wonder what a carrier in South America or Africa thinks about these accusations.

I have clients who have purchased electronics from these two Chinese companies. In the FTTP arena the two companies have highly competitive pricing, which is attractive to smaller ISPs updating their networks to fiber. Huawei also offers several upgrade solutions for HFC cable networks that are far less expensive than the handful of other vendors offering solutions.

The announcements by the US government creates a quandary for anybody who has already put this gear into their network. At least for now the potential problems from using this equipment have not been specifically identified. So a network owner has no way of knowing if the problem is only with cellphones, if it applies to everything made by these companies, or even if there is a political nature to these warnings rather than a technical one.

Any small carrier using this equipment likely cannot afford to remove and replace electronics from these companies in their networks. The folks I know using ZTE FTTP gear speak high praises of the ease of using the electronics – which makes sense since these two companies have far more installed fiber customers worldwide than any other manufacturer.

Somebody with this equipment in their network has several quandaries. Do they continue to complete networks that already use this gear or should they somehow introduce a second vendor into their network – an expensive undertaking. Do they owe any warnings to their own customers (at the risk of losing customers). Do they do anything at all?

For now all that is in place is a warning from US intelligence agencies not to use the gear, but there is no prohibition from doing so. And even should the Senate bill pass it would only prohibit ISPs using the gear from providing telecom services to military bases – a business line that is largely handled by the big telcos with nationwide government contracts.

I have no advice to give clients on this other than to strongly consider not choosing these vendors for future projects. If the gear is as bad as it’s being made to sound then it’s hard to understand why the US government wouldn’t ban it rather than just warn about it. I can’t help but wonder how much of this is international wrangling over trade rather than any specific threat or risk.

SDN Finally Comes to Telecom

For years we’ve heard that Software Defined Networking (SDN) is coming to telecom. There have been some movement in that area in routing on long-haul fiber routes, but mostly this network concept is not being used in telecom networks.

AT&T just announced the first major deployment of SDN. They will be introducing more than 60,000 ‘white box’ routers into their cellular networks. White box means that the routers are essentially blank generic hardware that comes with no software or operating systems. This differs from the normal routers from companies like Cisco that come with a full suite of software that defines how the box will function. In fact, from a cost perspective the software costs a lot more than the software in a traditional router.

AT&T will now be buying low-cost hardware and will load their own software onto the boxes. This is not a new concept and the big data center companies like Facebook and Google have been doing this for several years. SDN let’s a provider load only the software they need to support just the functions they need. The data center providers say that simplifying the software saves them a fortune in power costs and air conditioning since the routers are far more efficient.

AT&T is a little late to the game compared to the big web companies, and it’s probably taken them a lot longer to develop their own proprietary suite of cell site software since it’s a lot more complicated than switches in a big data center. They wouldn’t want to hand their cell sites over to new software until it’s been tested hard in a variety of environments.

This move will save AT&T a lot of money over time. There’s the obvious savings on the white box routers. But the real savings is in efficiency. AT&T has a fleet of employees and contractors whose sole function is to upgrade cell sites. If you’ve followed the company you’ve seen that it takes them a while to introduce upgrades into their networks as technicians often have to visit every cell site, each with different generics of operating hardware and software.

The company will still need to visit cell sites to make hardware changes, but the promise of SDN is that software changes can be implemented across their whole network in a short period of time. This means they can fix security flaws or introduce new features quickly. They will have a far more homogeneous network where cell sites use the same generics of hardware and software, which should reduce glitches and local problems. The company will save a lot on labor and contractor costs.

This isn’t good news for the rest of the industry. This means that Cisco and other router makers are going to sell far fewer telecom-specific routers. The smaller companies in the country have always ridden the coattails of AT&T and Verizon, whose purchase of switches and routers pulled down the cost of these boxes for everybody else. These big companies also pushed the switch manufacturers to constantly improve their equipment, and the volume of boxes sold justified the router manufacturers to do the needed R&D.

You might think that smaller carriers could also buy their own white box routers to also save money. This looks particularly attractive since AT&T is developing some of the software collaboratively with other carriers and making the generic software available to everybody. But the generic base software is not the same software that will run AT&T’s new boxes. They’ve undoubtedly sunken tens of millions into customizing the software further. Smaller carriers won’t have the resources to customize this software to make it fully functional.

This change will ripple through the industry in other ways. For years companies often hired technicians who had Cisco certification on various types of equipment, knowing that they understood the basics of how the software could be operated. But as Cisco and other routers are edged out of the industry there are going to be far fewer jobs for those who are Cisco certified. I saw an article a few years ago that predicted that SDN would decimate the technician work force by eliminating a huge percentage of jobs over time. AT&T will need surprisingly few engineers and techs at a central hub now to update their whole network.

We’ve known this change has been coming for five years, but now the first wave of it is here. SDN will be one of the biggest transformational technologies we’ve seen in years – it will make the big carriers nimble, something they have never been. And they are going to make it harder over time for all of the smaller carriers that compete with them – something AT&T doesn’t mind in the least.