Tag Archives: machine-to-machine

The Explosion of Machine-to-Machine Traffic

Fierce Networks recently reported that on the earnings call for the first quarter of this year, CEO Justin Hotard of Nokia said that machine-to-machine (M2M) Internet traffic will explode, and in a few years will become the largest source of data transmitted across the Internet. He said that about 20% of all network traffic today, about 80 exabytes, comes from machine-to-machine traffic, and that alone is big news. Nokia is betting its future growth will come from meeting this growing demand.

I went back and looked at older blogs, and in 2020, there were estimates that between 3% and 5% of Internet traffic came from machine-to-machine traffic. In 2020, it was estimated that video streaming accounted for 80% of the traffic on the Internet. I remember a lot of discussion around the industry at that time asking why we were spending so much money to expand broadband networks to support entertainment.

How did M2M traffic grow to be such a big part of the Internet today? There are a lot of different sources of M2M traffic, a few that should be familiar to most of us, but some that are not obvious:

  • Software Driven Usage. Compared to 2020, much of the software we use in homes and businesses has moved to the cloud. It’s not M2M traffic when you save a Word file in the cloud, but it is M2M traffic when the software you use communicates with the cloud behind the scenes.
  • Internet of Things. Consumer Affairs recently estimated that the average U.S. home now has 21 connected devices. That’s double the number of average devices from 2020. A large percentage of these devices transmit data to the cloud without the homeowner’s knowledge.
  • Industrial IoT. Businesses are a big source of M2M traffic. This includes video surveillance cameras and burglar alarms. It includes sensors of many kinds. Industrial IoT includes traffic involved with reading RFID tags, barcodes, and robotic scanners to track inventory and materials.
  • Automative. I think most people would be amazed at the large amounts of data transmitted from your vehicle if you have it connected to your home WiFi. This includes monitoring the sensors in the car as well as transmitting performance statistics every time you drive the car.
  • Healthcare Monitoring. There is a growing use of health monitors that track devices like pacemakers and glucose monitors. Patients are routinely being monitored after having medical procedures.

While the traffic from all of these sources is growing, the huge M2M growth predicted by Nokia will largely be driven by the use of AI. AI is currently driving a huge amount of new traffic. A lot of it is coming from AI datacenters that communicate with each other. There is also huge traffic growth coming on the edge from uses like AI-driven factories and other industrial uses that are using software to replace people.

Nokia sees its own growth coming from supplying the gear needed to provide faster transmission speeds on long-haul fiber and keeping up with the growth to power the growing number of long-haul fiber routes being built. Nokia also expects to see growth at the edge as the need to be able to process huge amounts of M2M traffic keeps increasing.

Interestingly, there is a big push among the companies that operate data centers to create a parallel long-haul network that segregates AI traffic from everything else. This is being done to improve latency by eliminating contention with other traffic and by eliminating intermediate traffic switching points. Data centers also want to use fiber routes with the fastest speeds and largest capacity.

Machine Generated Broadband

One of the more interesting predictions in the latest Cisco annual internet forecast is that there will be more machine-to-machine (M2M) connections on the Internet by 2021 than there are people using smartphones, desktops, laptops and tablets.

Today there are a little over 11 billion human-used machines connected to the Internet. That number is growing steadily and Cisco predicts that by 2021 there will be over 13 billion such devices using the Internet. That prediction also assumes that total users on the internet will grow from a worldwide 44% broadband penetration in 2016 to a 58% worldwide penetration of people that have connectivity to the Internet by 2021.

But the use of M2M devices is expected to grow a lot faster. There are fewer than 6 billion such devices in use today and Cisco is projecting that will grow to nearly 14 billion by 2021.

So what is machine-to-machine communication? Broadly speaking it is any technology that allows networked devices to exchange information and perform actions without assistance from humans. This encompasses a huge range of different devices including:

  • Cloud data center. When something is stored in the cloud, most cloud services create duplicate copies of data at multiple data centers to protect against a failure at any given data center. While this does not represent a huge number of devices when measured on the scale of billions, the volume of traffic between data centers is gigantic.
  • Telemetry. Telemetry has been around since before the Internet. Telemetry includes devices that monitor and transmit operational data from field locations of businesses, with the most common examples being devices that monitor the performance of electric networks and water systems. But the devices used for telemetry will grow rapidly as our existing utility grids are upgraded to become smart grids and when telemetry is used by farmers to monitor crops and animals, used to monitor wind and solar farms, and used to monitor wildlife and many other things in the environment.
  • Home Internet of Things. Much of the growth of devices will come from an explosion of devices used for the Internet of Things. In the consumer market that will include all of the smart devices we put into homes such as burglar alarms, cameras, smart door locks and smart appliances of many kinds.
  • Business IoT. There is expected to be an even greater proliferation of IoT devices for businesses. For example, modern factories that include robots are expected to have numerous devices that monitor and direct the performance of machines. Hospitals are expected to replace wires with wireless networked devices used to monitor patients. Retail stores are all investigating devices that track customers through the store to assist in shopping and to offer inducements to purchase.
  • Smart Cars and Trucks. By 2021 it’s expected that most new cars and trucks will routinely communicate with the Internet. This does not necessarily imply self-driving vehicles, but rather that all new vehicles will have M2M capabilities.
  • Smart Cities. A number of large cities are looking to improve living conditions using smart city technologies. This is going to require the deployment of huge numbers of sensors that will be used to improve things like traffic flow, monitoring for crimes and improvement everyday things like garbage collection and snow removal.
  • Wearables. Today there are huge numbers of fitness monitors, but it’s expected that it will become routine for people to wear health monitors of various types that keep track of vital statistics and monitor to catch problems at an early stage.
  • Gray Areas. There are also a lot of machine-to-machine communications that come from computers, laptops and smartphones. I see that my phone uses data even at those times when I’m not using it. Our devices now query the cloud to look for updates, to make back-ups of our data or to take care of other tasks that our apps do in the background without our knowledge or active participation.

Of course, having more machine-to-machine devices doesn’t mean that this traffic will grow to dominate web traffic. Cisco predicts that by 2021 that 83% of the traffic on the web will be video of some sort. While most of that video will be used for entertainment, it will also include huge piles of broadband usage for surveillance cameras and other video sources.

If you are interested in M2M developments I recommend M2M: Machine2Machine Magazine. This magazine contains hundreds of articles on the various fields of M2M communications.