Data centres and data transmission networks

Tracking Clean Energy Progress

🕐 Last updated Friday, 29 March 2019

On track

As the world becomes increasingly digitalized, data centres and data transmission networks are emerging as an important source of energy demand, each accounting for about 1% of global electricity demand. Despite exponential growth in demand for these services, sustained gains in energy efficiency could keep overall energy demand growth largely in check over the next five years.


Data centres

Global data centre electricity demand in 2017 amounted to an estimated 195 TWh, or about 1% of global final demand for electricity (Masanet et al., 2018). Despite a projected doubling of data centre IP traffic and workloads, global data centre energy demand is projected to remain flat to 2021 based on expected efficiency trends (Cisco, 2018; Masanet et al., 2018; Shehabi et al., 2016).

Global data centre energy demand

Despite an expected doubling of data centre workloads, global data centre electricity demand appears to be on track to remain flat in coming years.

	Infrastructure*	Network	Storage	Servers
2015	84.3	3.7	14.2	88.4
2017E	77.3	3.9	17.2	96.6
2021P	58.6	3.9	19.2	109.1
{
"chart":{"type":"column"},
"title": {
	"text": "By end use"
	},
"colors":["#00b050",
              "#92d050",
              "#31869b",
              "#93cddd",
              "#60497a",
              "#b1a0c7",
              "#da8137",
              "#c0504d",
              "#ffc000",
              "#ea9854"],
"xAxis":{"type":"category"},
"plotOptions":{"column":{"borderWidth":0,"stacking":"normal"}},
"tooltip":{"valueDecimals": 0, "valueSuffix": " TWh"},
"yAxis":{"min":0,"max":200,"title":{"useHTML":true,"text":"TWh"}}
}
	Hyperscale	Cloud (non-hyperscale)	Traditional
2015	31.1	62	97.6
2017E	49.8	75.1	70.1
2021P	86.6	71.6	32.6
{
"chart":{"type":"column"},
"title": {
	"text": "By data centre type"
	},
"colors":["#00b050",
              "#92d050",
              "#31869b",
              "#93cddd",
              "#60497a",
              "#b1a0c7",
              "#da8137",
              "#c0504d",
              "#ffc000",
              "#ea9854"],
"xAxis":{"type":"category"},
"plotOptions":{"column":{"borderWidth":0,"stacking":"normal"}},
"tooltip":{"valueDecimals": 0, "valueSuffix": " TWh"},
"yAxis":{"min":0,"max":200,"title":{"useHTML":true,"text":"TWh"}}
}

* Data centre infrastructure refers to energy consumed by non-IT equipment such as cooling and lighting.
Source: Masanet et al. (2018)

The strong growth in demand for data centre services is offset by continued improvements in the efficiency of servers, storage devices, network switches and data centre infrastructure, as well as a shift to much greater shares of cloud and hyperscale data centres.

The shift away from small inefficient data centres towards much larger cloud and hyperscale data centres, often through outsourcing data centre services, is a major and growing source of energy efficiency gains. This trend is evident in the shrinking share of data centre infrastructure in total energy demand.

Hyperscale data centres – very efficient, large-scale public cloud data centres – run at high capacity, thanks in part to virtualisation software that enables data centre operators to deliver higher work outputs with fewer servers. However, hyperscale data centres are only viable for operations where latency – the time delay in data transmission – is not critical, given that hyperscale data centres are typically located further away from the end user.

The nascent IT infrastructure for blockchain and cryptocurrencies is evolving rapidly, and its implications for data centre electricity use are not yet well understood.

Early estimates suggest that the electricity use of Bitcoin miners – one prominent example of the emerging blockchain IT infrastructure – may currently be on the order of around 0.1-0.3% of global electricity use (Bendiksen & Gibbons, 2018; Bendiksen, Gibbons, & Lim, 2018; Bevand, 2018; Bloomberg New Energy Finance, 2018; De Vries, 2018; Digiconomist, 2018; Krause & Tolaymat, 2018; Morgan Stanley, 2018; Vranken, 2017). However, as blockchain applications grow, understanding and managing its energy use implications may become increasingly important for the energy analysis and policy communities.


Data transmission networks

Data networks consumed around 185 TWh globally in 2015, or another 1% of total demand, with mobile networks accounting for around two-thirds of the total.

Over the near-term, the range of possible energy outcomes is wide, hinging largely on the growth of demand for data and the pace of further efficiency improvements. In a moderate efficiency-improvement scenario of 10% per year (based on conservative estimates of historical improvements), electricity demand could rise by over 70% to about 320 TWh.

Under the high efficiency-improvement scenario of 20% per year (based on the historic rates achieved in well-managed networks in developed countries with high capacity utilisation), the demand could drop by 15% to about 160 TWh. This large range of potential outcomes highlights the important role for policy to drive further efficiency gains.

Electricity use by data transmission networks

The prospects for future electricity demand from data transmission networks hinge critically on the pace of efficiency improvements.

	Fixed	Mobile
2015	66	118.5
2021 - Moderate efficiency improvement	117.5	200
2021 - High efficiency improvement	58	99
{
"chart":{"type":"column"},
"colors":["#00b050",
              "#92d050",
              "#31869b",
              "#93cddd",
              "#60497a",
              "#b1a0c7",
              "#da8137",
              "#c0504d",
              "#ffc000",
              "#ea9854"],
"xAxis":{"type":"category"},
"plotOptions":{
	"column":{"borderWidth":0,"stacking":"normal"},
	"series": {
		"events": {
			"legendItemClick": "function () { return false; }"
		}
	}
},
"tooltip":{"enabled":false},
"yAxis":{"reversedStacks": false, "min":0,"max":500,"title":{"useHTML":true,"text":"TWh"}},
"series": [{},{ "id": "fixed" }, {
	"type": "errorbar",
	"linkedTo": "fixed",
	"data": [ [254, 115], [439.5, 195.5], [217, 97] ]
	}]
}

Several relevant trends are shaping the future of data network electricity use (Cisco, 2017). Global IP traffic increased more than threefold over 2011-16 and is projected to grow by a similar rate between 2016 and 2021. The nature of data transmission is changing rapidly, with traffic from wireless and mobile devices expected to account for more than 63% of total IP traffic by 2021, up from 49% in 2016.

This shift toward greater use of mobile networks may have significant implications for the energy use of data transmission networks, given the considerably higher electricity intensities (kWh/GB) of mobile networks compared with fixed-line networks at current traffic rates. While the latest mobile telecommunications technologies are much less energy intensive than older technologies (e.g. 4G can be more than 50 times more energy efficient than 2G), their higher speeds may also allow for greater usage and traffic volumes.

Electricity intensity of network transmission by access type in 2015

Mobile networks typically have higher electricity intensities (kWh/GB) than fixed-line networks, but newer generation mobile networks and increased mobile network capacity utilisation are closing the gap.


{
	"chart": {
	  "type": "columnrange"
	},
	"yAxis": {
	  "title": {
	  	"text": "kWh/GB"
	  },
	  "type": "logarithmic",
	  "min": 0.01,
	  "max": 100
	},
	"tooltip": {
		"headerFormat": "",
		"valueSuffix": " kWh/GB"
	},
	"xAxis": {
	  "visible": false
	},
	"series": [
		{ 
		  "name": "Fixed",
		  "data": [ [ 0.04, 0.12 ] ]
		},
		{ 
		  "name": "Mobile 2G",
		  "data": [ [ 12.92, 24.44 ] ]
		},
		{ 
		  "name": "Mobile 3G",
		  "data": [ [ 1.03, 1.96 ] ]
		},
		{ 
		  "name": "Mobile 4G",
		  "data": [ [ 0.229, 0.45 ] ]
		}
	]
}

Despite strong growth in data demand and shifts to mobile transmission, three important trends may limit overall growth in energy demand:

  • Data transmission network technologies are rapidly becoming more efficient, e.g. fixed-line network energy intensity has halved every two years since 2000 in developed countries (Aslan et al., 2017). Mobile access network energy efficiency has in recent years improved at annual rates of around of 10-20% (Fehske et al. 2011).
  • Capacity utilisation of networks is increasing, driving down energy use per byte sent, even with existing equipment.
  • Mobile networks are shifting rapidly away from older networks towards more efficient 4G. By 2021, 4G is expected to cover around 80% of mobile traffic, while 2G is expected to cover less than 1% (Cisco, 2017).

Beyond the next five years, providing credible assessments of energy use by digital technologies is extremely difficult. Direct energy use over the long run will continue to be a battle between data demand growth versus the continuation of efficiency improvements. If demand for digital services outpaces efficiency gains, powering data centres and networks with renewable energy will become increasingly important to curbing greenhouse gas emissions.


References

  1. Aslan, J., Mayers, K., Koomey, J. G., & France, C. (2017). Electricity Intensity of Internet Data Transmission: Untangling the Estimates. Journal of Industrial Ecology. https://doi.org/10.1111/jiec.12630
  2. Bendiksen, C., & Gibbons, S. (2018). The Bitcoin Mining Network - Trends, Marginal Creation Cost, Electricity Consumption & Sources. https://s3-eu-west-2.amazonaws.com/assets.coinshares.co.uk/wp-content/uploads/2018/06/06140515/MiningWhitepaperFinal.pdf
  3. Bendiksen, C., Gibbons, S., & Lim, E. (2018). The Bitcoin Mining Network - Trends, Marginal Creation Cost, Electricity Consumption & Sources. https://coinshares.co.uk/wp-content/uploads/2018/11/Mining-Whitepaper-Final.pdf
  4. Bevand, M. (2018). Electricity consumption of Bitcoin: a market-based and technical analysis. http://blog.zorinaq.com/bitcoin-electricity-consumption/#summary
  5. Bloomberg New Energy Finance. (2018). Bitcoin Energy Crisis as China Cracks Down.
  6. Cisco. (2017). Cisco Visual Networking Index: Global Mobile Data Traffic Forecast Update, 2016-2021. http://www.cisco.com/c/en/us/solutions/collateral/service-provider/visual-networking-index-vni/mobile-white-paper-c11-520862.pdf
  7. Cisco. (2018). Cisco Global Cloud Index: Forecast and Methodology, 2016–2021. https://www.cisco.com/c/en/us/solutions/collateral/service-provider/global-cloud-index-gci/white-paper-c11-738085.pdf
  8. De Vries, A. (2018). Bitcoin’s Growing Energy Problem. https://doi.org/10.1016/j.joule.2018.04.016
  9. Digiconomist. (2018). Bitcoin Energy Consumption Index. https://digiconomist.net/bitcoin-energy-consumption
  10. Fehske, A., Fettweis, G., Malmodin, J., & Biczok, G. (2011). The global footprint of mobile communications: The ecological and economic perspective. IEEE Communications Magazine, 49(8), 55–62. https://doi.org/10.1109/MCOM.2011.5978416
  11. International Energy Agency. (2017). Digitalization & Energy. http://www.iea.org/publications/freepublications/publication/DigitalizationandEnergy3.pdf
  12. Krause, M. J., & Tolaymat, T. (2018). Quantification of energy and carbon costs for mining cryptocurrencies. Nature Sustainability, 1(11), 711–718. https://doi.org/10.1038/s41893-018-0152-7
  13. Masanet, E. R., Shehabi, A., Smith, S. J., & Lei, N. (2018). Global Data Center Energy Use: Distribution, Composition, and Near-Term Outlook (Northwestern University Working Paper). Evanston, IL.
  14. Morgan Stanley. (2018). Bitcoin ASIC production substantiates electricity use; points to coming jump. https://login.matrix.ms.com/public/login-email/webapp/entry?uri=/eqr/article/webapp/b974be48-effb-11e7-8cdb-6e28b48ebbd3
  15. Shehabi, A., Smith, S. J., Sartor, D. A., Brown, R. E., Herrlin, M., Koomey, J. G., … Lintner, W. (2016). United States Data Center Energy Usage Report. Berkeley, California. https://doi.org/LBNL-1005775
  16. Vranken, H. (2017). Sustainability of bitcoin and blockchains. Current Opinion in Environmental Sustainability, 28, 1–9. https://doi.org/10.1016/J.COSUST.2017.04.011