PresentationPDF Available

Total Consumer Power Consumption Forecast

Authors:

Abstract and Figures

This presentation outlines an estimation of the global electricity usage that can be ascribed to Communication Technology (CT) in the coming decade. The scope is two scenarios for use and production of consumer devices, communication networks and data centers. Two different scenarios— best and expected—are set up, which include annual numbers of sold devices, data traffic and electricity intensities/efficiencies. For the first time AR and VR devices will be included in a CT power trend analysis. I will emphasize the potential development of total Global Data Center IP Traffic and data center electric power consumption. The effect of 5G adaption rate on the total CT power consumption will also be addressed. The likely share of CT of the total global electric power consumption will be discussed in the light of CTs potential to reduce consumption.
No caption available
… 
No caption available
… 
No caption available
… 
No caption available
… 
No caption available
… 
Content may be subject to copyright.
Total Consumer Power
Consumption Forecast
by Dr. Anders S.G. Andrae (Huawei) at
the Nordic Digital Business Summit,
Helsinki, Finland, October 5, 2017
Introduction
Methodological approach + data trends
Emerging trends in electricity use of production of
hardware
Emerging trends in electricity use of consumer devices
Emerging trends in electricity use of fixed and wireless
networks
Emerging trends in electricity use of data centers
Synthesis of the global trend
Discussion
Concluding remarks
Outlook
Contents
Introduction
The are conflicting messages regarding the path to sustainability:
-different ways to measure
-different statistics
-we got to prepare ourselves for a truth about the environment we may not like?
-we might enter the YottaByte (1024 Byte) era in the next decade if so, can the
effect on power consumption be understood?
Problems with several existing ICT footprint investigations:
-too limited (geographical and temporal) system boundary
-overestimation of power saving potential in the next decade
-assume that historicallowpower use can predict future global power use in
the next decade withforeseenunprecedented data traffic growth
-assume that Moore´s law relation to digital circuitry can continue “forever
-”wrong” slicing of the networks and thereby possible double counting of e.g.
the core network
Can future consumer ICT infrastructure actually slow its overall electricity use
until 2025?
Under which circumstances is the power consumption of ICT slowing?
Under which circumstances is it rising?
Introduction ICTs share of the big
picture for electricity
Sources: Adaptions of Malmodin, J., & Bergmark, P. (2015). and Andrae&Edler (2015).
Industry
53.3%
Buildings
45.4%
Travel of people
0.5%
Transports of goods
0.8%
Electricity use in the world in 2015, ≈22000 TWh, ICT is ≈1700 TWh spread
in different Sectors
Introduction ICTs share of the big
picture for GHG
Source: Adaptions of Malmodin, J., & Bergmark, P. (2015).
Industry
37%
Buildings
21%
Travel of people
6%
Transports of goods
11%
Land use
10%
Waste
3%
Greenhouse gas emission in 2015 by Sector, ≈52 Gigatonnes, ICT is 1 Gt
spread in different Sectors
Methodological approach Global Scope
1.Consumer (“client+TV”) devices: Desktops, Monitors, Laptops,
Smartphones, Tablets, Ordinary Mobile Phones, Phablets, Mobile
Broadband Modems, TVs, TV peripherals (Set Top Boxes and
Game Consoles)+Smarthome devices+Wearable devices+AR/VR
devices
2.Wireless (mobile) Access Networks: radio base stations, mobile
switching
3.Fixed Wired Access Networks: Optical core wired access networks
4. Fixed Wi-Fi Access networks: Customer Premise Equipment+WLAN
5. Data Centers: Entire data centers including cooling
6. Production of all: Production from LCAs for devices, lifecycle ratio
method for Networks and Data Centers
Methodological approach trends
Trends are more important than ”exact” use patterns and numbers, as we do not
exactly know how and which devices will be used in the future.
Fundamental conflict (for wireless mobile) between cost, bandwidth efficiency and
energy efficiency.
Optimizing for energy efficiency (EE) will be essential going forward.
MAJOR TRENDS:
IoT
Artificial Intelligence
Augumented Reality
Virtual Reality
Fog computing, SDN, Virtualization
Mobile Edge Computing to increase battery life of mobile phones
High-frequency EE antennae, for exampel GAPWAVES
35% annual increase of the average peak data rate from 2G 5G and 6G
(expected in 2030 with at least 80Gb/s), 20 times higher rates for each G.
Global Data trends
9%
10%
12%
15%
18%
22%
25%
28%
31%
34%
37%
0%
5%
10%
15%
20%
25%
30%
35%
40%
2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025
%
Year
Share of mobile communication of daily
communication 2015-2025
0.4
0.4
0.5
0.6
0.8
1.0
1.2
1.6
1.9
2.5
3.1
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
2015 2017 2019 2021 2023 2025
GB/capita/day
Year
Daily communication (GB) per capita
2015-2025, "data center to user=access"
2
3
8
12
17
24
35
50
72
2
3
4
5
5
6
8
10
13
16
21
0
10
20
30
40
50
60
70
80
2015 2017 2019 2021 2023 2025
GB/capita/day
Year
Daily data generation (GB) per capita
2015-2025, "within, between+access"
Emerging trends in global electricity use
of production of hardware
Best case assumptions:
Lowest upstream LCA value for production in
2010,
5% annual improvement of productivity,
lowest value (10%) found for share of
manufacturing electricity, Networks and
Data Centers 10%
Expected case assumptions: 3%
improvement, 15% share of manufacturing
Source: Adaptions of the Supplementary Materials of http://www.mdpi.com/2078-1547/6/1/117
Smarthome devices, Wearable devices, and AR/VR
devices increase the electricity use of production by
≈5-10% .
The share of the Production of Data Center
Hardware and Network Hardware is expected to rise.
400
259
915
42
280
-200
0
200
400
600
800
1000
1200
2015 2017 2019 2021 2023 2025
Year
Electricity usage (TWh) of Production 2015-2025
Production Best
Production Expected
Million produced consumer
electronic devices×100
Others
9%
AR/VR devices
1%
Wearable devices
3%
Smarthome
devices
4%
Phablets
5%
Tablets
5%
Laptops
6%
TV
9%
Networks eq
10%
Smartphones
15%
Data center eq
33%
Share of global ICT production electricity
use by type in 2025
Emerging trends in global electricity use
of consumer devices Model includes:
Lifetimes
Units produced/year
Average electricity
used/unit/year
Annual reduction of
electricity (%)
The client device power use is dominated by
processing and displays.
Current downward trend is expected to continue.
Power saving (architectures, management,
technologies) will allow improvements.
Smarthome devices, wearables and AR/VR seem
to be ≈1% of total in 2025. TVs&peripherals >60%.
Source: Adaptations of Supplementary Materials of http://www.mdpi.com/2078-1547/6/1/117
744
961
801
486
100
600
1 100
2015 2017 2019 2021 2023 2025
Year
Global Electricity usage (TWh) of Consumer Devices 2015-2025
Consumer devices Use Best
Consumer devices Use Expected
Million devices in use, *100
TV
46.30%
Smarthome devices
0.80%
Wearable devices
0.36%
ARVR devices
0.06%
Laptops
15.17%
TV Set Top Boxes
8.80%
Desktops
7.57%
TV Game Consoles
4.72%
A/V Receiver
4.51%
Monitors
3.74%
Smartphones
3.23%
DVD/Blueray
2.27%
Tablets
1.33%
Phablets
1.14%
Share of consumer device global Electricity use by
type in 2025
Emerging trends in global traffic and
electricity use of fixed wired access
networks
Model includes:
Global electricity usage
in 2011 (178 TWh) and
2012 (196 TWh)
Fixed access wired +
fixed access Wi-Fi data
traffic in 2011 (390 EB/y)
and 2012 (470 EB/y) and
so on…
Annual electricity
intensity improvement
(EI), 22% (best) and 15%
p.a. (expected)
From 2022, for EI only,
5% is assumed possible
for both scenarios as I
expect it will become
more difficult to improve
the EI via Moore’s Law.
The large energy saving improvements from phasing out copper
are already included Future improvement rate will eventually not
mitigate traffic growth. Still, the average EI could be improved ≈5
times between 2015 and 2025.
Source: Adaptations of http://www.mdpi.com/2078-1547/6/1/117
226
167
339
81
633
0
100
200
300
400
500
600
700
2015 2017 2019 2021 2023 2025
Year
Global Electricity usage (TWh) of Fixed access wired networks 2015-2025
Fixed access wired Best
Fixed access wired
Expected
Average traffic
(ExaBytes/year)×10
Emerging trends in global traffic and electricity
use of fixed Wi-Fi access networks
Model includes:
Global electricity usage
of CPE (46 TWh) in 2011
and 2012 (51 TWh)
Fixed access Wi-Fi data
traffic in 2011 (154 EB/y)
and 2012 (200 EB/y) and
so on..
Annual electricity
intensity improvement (EI)
22.5% p.a. (best) and 15%
p.a. (expected)
From 2022, for EI only,
5% is assumed possible for
both scenarios.
CPEs (modems, gateways, ONTs) are improving their energy efficiency,
but the numbers are growing assumingly along the traffic. The Wi-Fi
backhaul infrastructure (included here) is also expected to keep
growing.
Source: http://www.mdpi.com/2078-1547/6/1/117
77
279
47
492
0
100
200
300
400
500
600
2015 2017 2019 2021 2023 2025
Year
Global Electricity usage (TWh) of Fixed access Wi-Fi networks 2015-2025
Fixed access WiFi Best
Fixed access WiFi
Expected
Average Wi-Fi traffic
(ExaBytes/year) ×10
Emerging trends in global traffic and electricity
use of wireless (mobile) networks
I currently expect a
reduction until around
2021 in global WAN due
to replacements
(”swapping”).
After 2022 I foresee an
increase of the electricity
use.
22% p.a. 2015-2021,
5% p.a. from 2022-2025
Should the
breakthrough of the
energy efficienct ”5G” be
delayed five years (1%
2025 instead of 1%
2020), an extra 47 TWh
could be used by WAN
globally.
KPN, between 2010 and 2015:
decreased their Network power consumption by 12%
improved their energy efficiency [GWh/Gbps] by 30% per
year for wireless and fixed networks.
I expect >50 times improvement of average EI globally.
Source: http://www.mdpi.com/2078-1547/6/1/117
118
137
220
176
31%
67%
5%
-50
0
50
100
150
200
250
2015 2020 2025
TWh, EB/year, %
Year
Wireless access electricity consumption from 2015 to 2025
Wireless (TWh) access
Expected
Average 4G data traffic
(EB/year)×10
Average 5G data traffic
(EB/year)×10
Share 4G of Wireless TWh, %
Share 5G of Wireless TWh, %
Source: KPN Integrated Annual Report 2015 Tables 2 and 6 in
Appendix
Emerging trends in global traffic and
electricity use of data centers
Model includes:
Global electricity
usage in Data Centers in
2010 (≈189 TWh) and
2011 (≈193 TWh).
Global Data Center IP
traffic in 2010 (≈1.4ZB)
and 2011 (≈1.8ZB).
Annual electricity
intensity improvement
(EI), 22.5% p.a. (best),
15% p.a. (expected).
From 2022, for EI
only, 5% is assumed
possible for all
scenarios.
Source: KPN Integrated Annual Report 2015 Table 2 in Appendix
1 204
203
3 390
5
10
29 (Cisco 15)
178
-500
0
500
1 000
1 500
2 000
2 500
3 000
3 500
4 000
2015 2017 2019 2021 2023 2025
Year
Electricity usage (TWh) of Data Centers 2015-2025
Data Centers Best
Data Centers Expected
Average Global Data Center IP
Traffic (ZettaBytes/year)
Between 2010 and 2015, KPNs data centers increased their
power consumption by 9%.
Synthesis: The global power trend for
ICT 2015 to 2025
2 788
1 757
5 860
0
500
1 000
1 500
2 000
2 500
3 000
3 500
4 000
4 500
5 000
5 500
6 000
2015 2017 2019 2021 2023 2025
Year
Global Electricity footprint (TWh) of Communication Technology
2015-2025
Best Case
Expected Case
Increasing energy efficiency including
enabling effect of SMART ICT Solutions
1756
4620
26475 (in line with
International Energy
Agency)
8524
0
5000
10000
15000
20000
25000
30000
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
Global electricity use (TWh) 2015-2025
ICT Sector own Electricity use, Best Case
ICT Sector Electricity use, Expected Case
Electricity use reduction by Smart ICT and
other measures
Total global electricity use, Best Case
Total global electricity use, Expected Case
Renewable electricity use
The share of ICT of global electricity usage: 2015 to
2025 with and without high global energy efficiency
gains
7.7%
6.6%
9.0%
10.5%
17.8%
8.2%
20.7%
0%
5%
10%
15%
20%
25%
2015 2017 2019 2021 2023 2025
Year
Communication Technology share of global electricity usage
Best Case "without"
considerable general energy
efficiency gains
Best case "with" EE, 420
TWhrs saved in 2015, 840 in
2016..and 4620 TWh in 2025
Expected Case "without" EE
Expected Case "with" EE
The global trends for ICT electricity
intensities 2015 to 2025
0.16
0.003
0.21
0.04
0.02
0.038
0.007, Data Centers
0.00
0.05
0.10
0.15
0.20
0.25
2015 2017 2019 2021 2023 2025
kWh/GigaByte
Year
Development of global electricity intensities in ICT networks
from 2015 to 2025
Electricity intensity wireless access
(kWh/GB)×10, Expected case
Electricity intensity fixed wired access
(kWh/GB), Best case
Electricity intensity fixed access Wi-Fi
(kWh/GB), Best case
Electricity intensity data centers
(kWh/GB), Best case
The global power repartition trends for
ICT between 2015 and 2025
0
500
1000
1500
2000
2500
3000
TWh/year
Best case scenario
Production
Data centers use
Wireless networks access use
Fixed access WiFi use
Fixed access wired use
Consumer devices use
0
1000
2000
3000
4000
5000
6000
7000
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
TWh/year
Expected case scenario
Production
Data centers use
Wireless networks access
use
Fixed access WiFi use
Fixed access wired use
Consumer devices use
The share of different sections of ICT of
global electricity use in 2015 and 2025
4.3%
0.7%
0.2%
0.5%
0.9%
1.2%
2.8%
0.9%
0.3%
0.5%
4.5%
1.5%
Consumer devices use
Fixed access wired use
Fixed access WiFi use
Wireless networks
access use
Data centers use
Production
Share of different ICT Sectors of global electricity 2015-2025,
Best case
Share of global electricity
usage in 2015
Share of global
electricity usage in 2025
How does this study compare to 1st global study
by Prof. Peter Corcoran in 2013?
725
40
502
403
375
2045
925
93
195
223
265
1701
Consumer
devices use
Wireless
Networks
Use
Fixed
Networks
Use
Data Centers
Use
Production
TOTAL
Comparison to earlier study for 2017
-Best case (TWh)
2017 Corcoran (2013) Best
Andrae (2017) Best
1115
59.6
671
541
408
2794
945
93
228
313
293
1872
Consumer
devices use
Wireless
Networks
Use
Fixed
Networks
Use
Data Centers
Use
Production
TOTAL
Comparison to earlier study for 2017
- Expected case
2017 Corcoran (2013) Expected
Andrae (2017) Expected
Other studies expected trends: Fixed and wireless
electricity use in Japan
Increasing electricity use is expected in
Japans telecom networks from now until
2030. Data centers, CPEs, production and
client devices are not seemingly included.
Sources: Ishii, K.; Kurumida, J.; Sato, K.-i.; Kudoh, T.; Namiki, S. Unifying Top-Down and Bottom-Up
Approaches to Evaluate Network Energy Consumption. Journal of Lightwave Technology 2015, 33, 4395.
Kishita, Y.; Yamaguchi, Y.; Umeda, Y.; Shimoda, Y.; Hara, M.; Sakurai, A.; Oka, H.; Tanaka, Y. Describing
Long-Term Electricity Demand Scenarios in the Telecommunications Industry: A Case Study of
Japan. Sustainability 2016, 8, 52.
Discussion
The speed of electricity intensity reduction vs. the speed of data traffic
increase.
Highly variable outlooks for the future power consumptions depending on
starting values” and percentual estimations of electricity intensity reductions
and data traffic increase.
Rebound effects new consumption of goods and services
“Ultra-efficient” Hong Kong's annual electricity consumptionand GHG
emissions from electricity consumptionis predicted to increase by 2030
Concluding Remarks
Future consumer ICT infrastructure cannot slow its overall electricity use until
2025.
The electric power consumption of the present ICT scope will be very significant
unless great efforts are put into power saving features.
It seems though that planned power saving measures and innovation will be able
to keep the electricity consumption of ICT and the World under control.
Outlook Different ”slicingof the framework
Production of Networks hardware
number of servers, routers, base stations,
modems,
Production of consumer devices.
Consumer devices use power,
Core network use power,
Access network use power,
Private data center use power,
Shared data center use power
might lead to different absolute values
and trends than my approach?
Widening the scope:
EXIOBASE: Resource extractions and
emissions related to the ICT industry.
Sustainability risk estimations: EPS2015
for expressing the sustainability costs and
savings.
Land use change
Biodiversity
Source: http://www.exiobase.eu/
Thanks for your attention!
anders.andrae@huawei.com
The global power repartition trends for
ICT between 2015 and 2025 (II)
Consumer devices use
55%
Networks Use
19%
Data Centers Use
11%
Production 20%
Best&Expected Case Scenario 2015
Consumer devices use
Networks Use
Data Centers Use
Production
Consumer
devices use
27%
Networks Use
16%
Data Centers Use
43%
Production 14%
Best Case Scenario 2025
Consumer devices use
Networks Use
Data Centers Use
Production
Consumer devices
use
14%
Networks Use
13%
Data Centers Use
58%
Production
15%
Expected Case Scenario 2025
Consumer devices use
Networks Use
Data Centers Use
Production
... Although AI has a great potential to minimize consumption and make grid-related efficiency optimized, it will still be a major consumer of electricity. According to research, data centers now require more than 2% of the world's electricity (Pearce, 2018), and scientists predict that, by 2025, this amount is expected to grow between 8% and 21% (Andrae and Edler, 2015;Andrae, 2017;Giles, 2019). A study by Belkhir and Elmeligi (2018) indicates that the estimated global footprint in 2020 may be compared to the impact of the aviation industry and greater than that of Japan (the fifth largest pollutant in the world). ...
Article
Full-text available
The aim of the article is to study the role of artificial intelligence (AI) in solving current issues of climate change, environmental protection and natural resources management. The advantages and threats of using AI for the development of political and legal parameters for ensuring the safe and effective implementation of technological system, as well as ensuring sustainable control over its functioning and development trends, are analyzed. The relevance of the topic is substantiated by the fact that the legislative basis in this area is at the early stage of formation, while the scale of the impact of AI on all the aspects of social life may be impossible to accurately foresee. A special attention is paid to the analysis of the legal regulation of these issues in the context of European Union and Ukraine. The present work is one of the few that addresses three issues: climate change, the growing influence of artificial intelligence, and the possibility of legal regulation of the use of AI to solve urgent environmental problems without threatening the fundamental human rights and freedoms.
... These impacts even induced public discussions about the necessity to introduce an unconditional basic income (Die Zeit 2015). • Planet: The billions of connected devices, communication infrastructures, and data centers require electricity in their usage and resources and energy for their production (Andrae 2017). The greenhouse gas emissions produced by these processes are comparable to those of air traffic today. ...
Chapter
Full-text available
To achieve long-term business success, purpose for associates, and acceptance in today’s societies, business leaders must keep a close eye on social and environmental dimensions in addition to the prerequisite of attaining adequate profits. This is especially true during times of major technological transformations. A fascinating breakthrough technology pushing such a transformation is artificial intelligence (AI). AI has great potential to make devices intelligent in the Internet of Things (IoT)—referred to as AIoT. In this chapter, we evaluate how a sustainable AIoT supports the three interrelated dimensions of profit, people, and planet.
... For example, ICT technology can be used to forecast energy consumption. Andre describes the case of total consumer power forecasting using ITC methods [69]. ...
Article
Full-text available
Leadership competencies are of crucial importance in every organisation as to a large extent they determine its success. This is especially evident in the time of Industry 4.0. Given this fact, the aim of our paper is to examine the relationship between leadership competencies and 4.0 leadership effectiveness. The heat and power plants industry was chosen as the subject of our research. The fuzzy-set qualitative comparative analysis (fs/QCA) was used as the research method. It enabled us not only to analyse particular variables, competences, and typical statistical relations between them, but we also revealed the patterns of causal relationships between particular variables. The key finding of our research was the juxtaposition of leadership competencies that are indispensable for 4.0 leaders in the CHP plants. We also found out that managerial competencies were not sufficient, and they should be supported by intellectual or socio-emotional ones.
... But still, it remains a major concern area. According to a survey; in year 2015, a Data Center (DC) consumed approximately 20% of all available electricity in the world [1] and this percentage is likely to reach 3000 TW hours by the end of year 2025 [2]. The reason behind this high energy consumption is establishment of more DCs to ensure better data availability and crash recovery. ...
Article
Full-text available
Efficient resource usage and providing quality of services to cloud users is challenging. Dynamic consolidation of virtual machines allows service provider to ensure better resource utilization that leverage virtual machine live migration strategy. However, excessive consolidation of virtual machines sometime deteriorates quality of services. An important issue for cloud service provider is to ensure energy-efficient virtual machine consolidation to optimize resource allocation keeping service-level agreement intact. Due to variability in workload, an adaptive policy must be ensured that adjusts the behaviour accordingly. This study presents an adaptive threshold policy that calculates upper threshold value taking into account past CPU utilization of host. The policy implements a location-free explicit estimator \(Q_n\) to analyze previous data periodically and adjust threshold accordingly. Performance comparison of proposed policy is done with existing estimators median absolute deviation and \(S_n\) and significantly improved results are obtained.
... networking, Google searches, or video viewing are growing every day in the world [1]. Due to this intensive use of software, ICT energy consumption had already increased by 2018 to 1,895 TWh, representing about 9% of total global energy consumption [2]. And by 2025, this could exceed 20% of total energy and emit up to 5.5% of the world's carbon emissions [3,4]. ...
Article
Software is a fundamental part of today's society. However, both users and software professionals need to be aware that its use impacts on the environment, due to the high energy consumption it entails. One of the main gaps to be faced is the difficulty of analyzing software energy consumption in the endeavor to know whether a particular software product is as much energetically efficient as possible, or at least more efficient than another, and to improve the environmental objectives of the software. For this reason, a Framework for Energy Efficiency Testing to Improve eNviromental Goals of the Software (FEETINGS) is presented in this paper. FEETINGS is composed of three main components: an ontology to provide precise definitions and harmonize the terminology related to software energy measurement; a process to guide researchers in carrying out the energy consumption measurements of the software, and a technological environment, which allows the capture, analysis and interpretation of software energy consumption data. This paper also presents an example of the application of the FEETINGS, which aims to raise awareness of the energy consumed by the software in activities that we perform daily, such as writing a tweet or a Facebook post. As a result, we have been able to verify that FEETINGS allows us to carry out an analysis and measurement of software energy consumption to provide users with good practices, as using an emoji or a picture rather than a GIF. © 2001 Elsevier Science. All rights reserved
... A data center can be a building or part of a building where data are gathered, processed, and stored [6]. According to a report by the Synergy Research Group, by the end of the third quarter of 2019, there were 504 hyperscale data centers worldwide [7], and another report predicted that this number would increase by 12-14% annually over the next five years [8]. Of all the data centers in 2019, approximately 40% reside in the US, and China, Japan, the UK, Germany, and Australia collectively contain 32%. ...
Article
Full-text available
An efficient cooling system for data centers can boost the working efficiency of servers and promote energy savings. In this study, a laboratory experiment and computational fluid dynamics (CFD) simulation were performed to explore the performance of a two-phase cooling system. The coefficient of performance (COP) and partial power usage effectiveness (pPUE) of the proposed system was evaluated under various IT (Information Technology) loads. The relationship between the interval of the two submerged servers and their surface temperatures was evaluated by CFD analysis, and the minimum intervals that could maintain the temperature of the server surfaces below 85 °C were obtained. Experimental results show that as server power increases, COP increases pPUE decreases. In one experiment, the COP increased from 19.0 to 26.7, whereas pPUE decreased from 1.053 to 1.037. The exergy efficiency of this system ranges from 12.65% to 18.96%, and the tank side accounts for most of the exergy destruction. The minimum intervals between servers are 15 mm under 1000 W of power, 20 mm under 1500 W, and more than 30 mm under 2000 W and above. The observations and conclusions in this study can be valuable references for the study of cooling systems in data centers.
Article
Thermal analysis of data centers is in urgent need to ensure that computer chips remain below the critical temperature while the energy consumption for cooling can be reduced. It is difficult to obtain detailed hotspot locations and temperatures of chips in large data centers containing hundreds of racks or more by direct measurement. In this paper, a multi-scale thermal analysis method is proposed that can predict the temperature distribution of chips and solder balls in data centers. The multi-scale model is divided into six scales: room, rack, server, Insulated-Gate Bipolar Transistor (IGBT), chip and solder ball. A concept of sub-model is proposed and the six levels are organized into four simulation sub-models. Sub-model 1 contains Room, Rack and Server (RRS); Sub-model 2 contains Server and IGBT (SI); Sub-model 3 contains IGBT and Chip (IC), and Sub-model 4 contains Chip and Solder-ball (CS). These four sub-models are one-way coupled by transmitting their results as boundary conditions between levels. The full-field simulation method is employed to verify the efficiency and accuracy of multi-scale simulation method for a single-rack data center. The two simulation results show that the highest temperature emerges in the same location. The Single-rack Full-field Model (SRFFM) costs 2.5 times more computational time than that with Single-rack Multi-scale Model (SRMSM). The deviation of the highest temperature of chips and solder balls are 1.57°C and 0.2°C between the two models which indicates that the multi-scale simulation method has good prospect in the data center thermal simulation. Finally, the multi-scale thermal analysis method is applied to a ship data center with 15 racks.
Article
Full-text available
Over the last years, increasing attention has been given to creating energy-efficient software systems. However, developers still lack the knowledge and the tools to support them in that task. In this work, we explore our vision that non-specialists can build software that consumes less energy by alternating diversely-designed pieces of software without increasing the development complexity. To support our vision, we propose an approach for energy-aware development that combines the construction of application-independent energy profiles of Java collections and static analysis to produce an estimate of in which ways and how intensively a system employs these collections. We implement this approach in a tool named CT+ that works with both desktop and mobile Java systems and is capable of analyzing 39 different collection implementations of lists, maps, and sets. We applied CT+ to seventeen software systems: two mobile-based, twelve desktop-based, and three that can run in both environments. Our evaluation infrastructure involved a high-end server, two notebooks, three smartphones, and a tablet. Overall, 2295 recommendations were applied, achieving up to 16.34% reduction in energy consumption, usually changing a single line of code per recommendation. Even for a real-world, mature system such as Tomcat, CT+ could achieve a 4.12% reduction in energy consumption. Our results indicate that some widely used collections, e.g., ArrayList, HashMap, and Hashtable, are not energy- efficient and sometimes should be avoided when energy consumption is a major concern.
Chapter
Das Thema Digitalisierung ist inzwischen in aller Munde, und das nicht erst seitdem die globale Corona-Pandemie in den zentralen Bereichen Arbeit und Bildung eine Digitalisierung in kürzester Zeit notwendig gemacht hat. Nicht umsonst galten Digitalisierung bzw. Konnektivität schon vorher als zentrale Megatrends. Obwohl insbesondere die Bedeutung des Themas für das eigene Leben und die wirtschaftliche Zukunft offensichtlich ist, sind viele Begriffe zum Thema, wie Internet 4.0, Big Data oder Internet of Things, zahlreichen Bundesbürgern noch nicht bekannt. Fahrerlose Verkehrsmittel wie der Skytrain am Frankfurter Flughafen, Selbstscan-Kassen in Supermärkten oder öffentlichen Bibliotheken, Mitfahrportale wie „Uber“ oder Sharing-Plattformen wie „AirBnB“ sowie zunehmende Onlinezeiten und zum Teil verödende Innenstädte zeigen die Veränderung von Leben und Arbeiten durch digitale Technologien. Der Artikel untersucht ausgewählte ökologische und soziale Chancen und Risiken aus diesen Entwicklungen und die sich daraus ergebenden politischen und gesellschaftlichen Implikationen.
Article
A novel method for estimating energy consumption of networks at the national scale is presented. Accurate estimations of network energy consumption are increasingly important because of continuously increasing network traffic and the need to assess emerging energy technologies to achieve overall energy efficiency. However, direct estimations are not practical because of the complex, widely diverse, and undisclosed configurations of actual networks. The impact of each network's energy overhead, which is the energy consumed above that required to accommodate network traffic, on the energy consumption associated with these complex configurations must be inferred. In the proposed method, the energy overhead is quantitatively evaluated and introduced as overhead factors by comparing the energy consumption estimated from network configuration models (bottom-up methods) with reports of actual energy consumption (top-down methods). This proposed 'unified' method is capable of long-term predictions of future technology trends, including network architecture changes. In this paper, we consider different overhead factors that serve as fitting parameters for different network areas, and demonstrate the procedure for determining each of these parameters through an estimation of the fixed broadband Internet in Japan. This unified method consistently estimates the long-term evolution of energy consumption from 2000 to 2030.