Our government and energy industry are revved up about the hoped for energy saving opportunities the IoT may provide. Armed with an astronomical amount of data that will be generated by the IoT, industry expects it will be better equipped to 1) reduce peak load on the grid; 2) design new energy saving technologies; and 3) implement energy-saving programs to reduce consumption or shift load to off-peak times. But what these projections fail to include in their analysis is the mega energy footprint of the IoT itself. It is not at all clear that the IoT will ever succeed in offsetting its own fast growing and unbounded energy consumption.
According to IEEE Consumer Electronics Magazine Editor Peter Corcoran, energy is one of three main determinants of the “longterm sustainability of the Internet of Things” (the other two being privacy and cyber security). In a recent article characterizing Corcoran as a “longterm IoT skeptic,” author Steven Max Patterson reports:
IoT devices are expected to be low-power devices, but the number of IoT devices that Cisco predicts will be 50 billion by 2020 is an order of magnitude larger than the number of smartphones and tablets in use today. If the energy consumed by these devices and the networks and data centers to which they are connected is considered, energy consumption by IoT will impactfully increase the rate of energy consumption growth.
Wireless technology consumes far more energy than fiber, cable or DSL
Wireless technologies consume far more energy than do wired technologies. According to Kris De Decker Why We Need a Speed Limit to the Internet, a wired connection is the most energy-efficient way to communicate digitally. If connection is made through a cellular network, energy use “soars.” According to a 2015 publication put out by the Centre for Energy Efficient Telecommunications,
“Our energy calculations show that by 2015, wireless cloud will consume up to 43 TWh, compared to only 9.2 TWh in 2012, an increase of 460%. This is an increase in carbon footprint from 6 megatonnes of CO2 in 2012 to up to 30 megatonnes of CO2 in 2015, the equivalent of adding 4.9 million cars to the roads. Up to 90% of this consumption is attributable to wireless access network technologies, data centres account for only 9%.”
“… the final link between telecommunications infrastructure and user device is by far the dominant and most concerning drain on energy in the entire cloud system . . . wireless access networks are clearly the biggest and most inefficient consumer of energy in the cloud environment.”
De Decker explains that 3G technologies use about 15 times more energy than wired connections, and 4G technologies consume 23 times more energy. There is no data yet on 5G. If the bulk of our Information Communications Technology (ICT) infrastructure and transmissions were wired, such as Fiber to the Home (FTTH), the energy footprint from our “digital world” would be significantly reduced. But because the IoT will greatly grow our dependence on wireless technologies, and because wireless consumes far more energy than wired alternatives, we are now facing an unprecedented exponential growth in our energy consumption.
(For more information on wireless consuming more energy than wired, please see http://whatis5g.info/microwave-radiation/#wireless)
Energy efficiency measures offset by usage
Digital technologies are becoming increasingly more energy efficient, and this trend will hopefully continue, at least for a while. But at the same time that we are making great strides in energy efficiency, more and more people are conducting greater portions of their lives online and spending greater amounts of time online. Furthermore, not all online activities consume energy equally — there is a hierarchy in energy consumption: The written word is the least energy intensive. Images consume more energy. And, to date, by far the most energy-intensive online activity is watching videos and particularly, high definition videos which 5G is promising. Online videos occupy an increasingly greater percentage of all online activity – restated, people are watching far more videos than ever before.
It is a common occurrence in digital technologies that improvements in energy efficiency of a particular device, bring about accompanying changes in usage of the device that often offset the hoped for benefits from energy savings. This is known as the “rebound effect”. For example, a smart phone uses significantly less energy than does a desktop computer. But due to the size of a smart phone, it can be carried around and used 24/7. This increase in use negates, or significantly reduces, energy savings.
We may presume that as more and more “things” in our world take up residence in the Cloud, people’s time online to connect with this data and their “digital selves” will also increase. Bryon Walsh, Senior Editor at TIME, writes,
“As our lives migrate to the digital cloud — and as more and more wireless devices of all sorts become part of our lives — the electrons will follow. And that shift underscores how challenging it will be to reduce electricity use and carbon emissions even as we become more efficient.”
FOUR WAYS THE IoE CONSUMES ENERGY
1. Data centers:
In an IoT world, data collected from billions of machines, appliances, “things,” and devices, as well as from sensors integrated into our environment, will be stored and responded to in data centers.
The 2014 paper, Data Center Efficiency Assessment, by the National Resources Defense Council (NRDC) reports that the 2 million computer servers in close to 3 million data centers that are used in the US for online activities “gulp enough electricity to power all of NYC’s households for 2 years.”
The NRDC goes on to explain:
“Data center electricity consumption is projected to increase to roughly 140 billion kilowatt-hours annually by 2020, the equivalent annual output of 50 power plants, costing American businesses $13 billion annually in electricity bills and emitting nearly 100 million metric tons of carbon pollution per year.” Please note these figures were compiled well before the IoT.
Greenpeace International explains it this way : “If the Cloud were a country it would have the fifth largest energy demand in the world.”
Although plans are underway to build future data centers in places where cleaner energy sources are available, the fact remains that data centers already consume mega amounts of energy, and this is only going to increase with the IoT.
2.Energy consumed from machine-to-machine (m2m) communications:
Machine-to-machine communications refers to 1) transmission of data from all Internet connected “things,” 2) remote software updates for personal devices, and 3) back-up of data, digital photos, and videos to the Cloud.
Hazas et al notes in Are there limits to growth in data traffic?, that although many m2m communications use very little energy, some, such as driverless cars or wearable medical devices, are highly data intensive and require inordinate amounts of energy, both at the source, and in the Cloud.
Hazas et al explain further that while each machine or “thing” may not consume a lot of energy in a given communication, the sheer volume of transmissions could place m2m communications as a top contender in energy consumption:
“This [machine to machine] communication will occur transparently, without observation or interaction, and potentially without limit. At the time of writing,  the existing 6.4bn connected IoT devices is only slightly less than world population (86%), but market predictions suggest this will reach 21bn [billion] devices by 2020— roughly three times world population estimates. Some predictions put machine-to-machine communication as 45% of the whole Internet traffic by 2022.”
3. Embodied energy: (Embodied energy, also known as “emergy,” refers to energy consumed in the production of goods. This includes the mining, manufacturing, transporting, and delivery of a product.)
Digital technology requires far more energy in the manufacturing process than do other products. Kris De Decker explains in The Monster Footprint of Digital Technology, that machines such as cars or refrigerators use far more energy during their “lifetime” than the amount used to manufacture them. According to De Decker, advanced digital technology has turned this relationship “upside down.” As counterintuitive as this may sound, he explains:
“A handful of microchips can have as much embodied energy as a car. And since digital technology has brought about a plethora of new products, and has also infiltrated almost all existing products, this change has vast consequences.”
Our new IoT technology will require billions of sensor nodes, and microchips each of which will require thousands of semiconductors. According to De Decker, the energy needed to produce one semiconductor is, “up to 6 orders of magnitude [emphasis added] above those of conventional manufacturing processes.”
In addition to the billions of Internet connected “things,” FCC Chair Wheeler projects that in a “5G world”, each person will own about 5-6 wireless devices. Add to that the millions of new small cells being deployed to accommodate all the digital “traffic” of the IoT, and it becomes evident that it will take an inordinate amount of energy just to produce the IoT.
4. Obsolescence of digital technologies
Perhaps the largest player in energy consumption of the IoT is that of (planned) obsolescence of all our technologies. Digital technologies generally need to be replaced every 1-3 years as Information and Communication Technologies evolve at an exceedingly fast pace. The fact that our digital devices have such a short lifespan, exacerbates the problem of the excessive amount of energy used in their production. The all too familiar need to “upgrade” will become yet more of a drain on our energy as literally billions of connected devices, machines, and “things“ will recurrently become obsolete and be discarded.
DeDecker maintains that, “Addressing technological obsolescence would be the most powerful approach to lower the ecological footprint of digital technology.”
Will our grid be able to absorb the IoT energy footprint?
Our energy grid was built over the course of the last century and is based on a business model that generates and delivers energy from burning fossil fuel. However, local renewable energy, such as rooftop solar is becoming increasingly more affordable and therefore increasingly able to replace coal and other polluting sources of energy. Utilities are trying desperately to stay afloat while integrating unprecedented amounts of local renewable energy into our saturated and failing energy grid. Gretchen Bakke explains in The Grid, The Fraying Wires Between Americans and our Energy Future, how challenging it is for our energy industry to integrate local renewables into the grid, while still providing reliable energy delivery and service to customers. Bakke gives the analogy of trying to rebuild an entire airplane fleet while keeping all the airplanes midflight. Faced with the voracious appetite of the wireless industry, our utilities will have an even greater challenge navigating the tenuous, but necessary path forward to a sustainable energy future.
Will this impact climate change?
An even stronger reason that energy consumption of wireless, 5G, and the IoT must be considered, is that our energy grid is still run primarily on fossil fuels – coal, natural gas and petroleum. So the exorbitant amount of energy that will be needed to support the IoT and its highly energy-intensive footprint will necessarily result in a huge increase in carbon emissions, thereby adversely impacting our climate. Will the energy footprint of 5G and the IoT be offset by improvements in energy efficiency and innovations businesses may come up with? Not certain at all.
For more on energy consumption and the IoT, please see,Tsunami of data’ could consume one fifth of global electricity by 2025