Energy

The energy usage of end-systems, particularly those that are within the control of users, have great potential for energy savings. As users and end-systems out-number datacentres by several orders of magnitude, even small savings per-user or end-system could have a huge global impact on energy savings.

The work below looks at several aspects of energy usage from the point of users, taking a "whole systems" approach to energy usage, and looking at the role of users and end-systems in overall energy consumption by users. Particularly, the relationship between energy usage savings and the impact on users needs to be understood so that users can make informed decisions on energy usage.

IEEE 802.11 / WiFi: examines the energy usage of client systems that communicate using IEEE 802.11 / WiFi.
Video: measurement and analysis of energy usage in video encoding / decoding.
Behaviour: the behaviour of users, end-systems and services in every-day usage.

(Some papers appear in more than one section below.)

IEEE 802.11 / WiFi

IEEE 802.11 (aka WiFi) is widely used, and its use is likely only to increase. The work below considers performance and energy usage within different application contexts.

M. Abu-Tair, S. N. Bhatti. Impact of cell load on 5GHz IEEE 802.11 WLAN. PAEWN 2017 - 12th Intl. Wkshp. Performance Analysis and Enhancement of Wireless Networks. Taipei, Taiwan. Mar 2017.
| PDF | .bib | 10.1109/WAINA.2017.27 | abstract We have conducted an empirical study of the latest 5GHz IEEE 802.11 wireless LAN (WLAN) variants of 802.11n (5GHz) and 802.11ac (Wave 1), under different cell load conditions. We have considered typical configurations of both protocols on a Linux testbed. Under light load, there is no clear difference between 802.11n and 802.11ac in terms of performance and energy consumption. However, in some cases of high cell load, we have found that there may be a small advantage with 802.11ac. Overall, we conclude that there may be little benefit in upgrading from 802.11n (5GHz) to 802.11ac in its current offering, as the benefits may be too small. M. Abu-Tair, S. N. Bhatti. Introducing IEEE 802.11ac into existing WLAN deployment scenarios. WiOpt 2015 - 13th Intl. Symp. Modeling and Optimization in Mobile, Ad Hoc, and Wireless Networks (WiNMeE 2015 - IEEE Intl. Workshop on Wireless Networks: Measurements and Experimentation). Mumbai, India. May 2015.
| PDF | .bib | 10.1109/WIOPT.2015.7151029 | abstract In mature wireless LAN (WLAN) deployments, we show that introducing 802.11ac could have little benefit compared to existing 802.11n deployments. Using a testbed with common characteristics for an existing WLAN deployment (such as an office environment), we compare throughput for 802.11ac and 802.11n (in both 5GHz and 2.4GHz bands). We find that 802.11ac has lower throughput than for 802.11n for our testbed configuration. We also provide an evaluation of energy usage for 802.11ac and 802.11n. M. Abu-Tair, S. N. Bhatti. Upgrading 802.11 deployments: a critical examination of performance. AINA 2015 - IEEE 29th Intl. Conf. Advanced Information Networking and Applications. Seoul, Korea. Mar 2015.
| PDF | .bib | 10.1109/AINA.2015.278 | abstract The increased demand for communications and Internet access makes Wireless Local Area Networks (WLANs) one of the most popular solutions for network connectivity. In this paper, we examine the performance and the energy efficiency of WLANs in 2.4 GHz and 5 GHz and discuss paths for upgrading. Our results show that it is worth upgrading to the 5 GHz bands from the 2.4 GHz band for 802.11n, especially for applications that are sensitive to packet loss. We also show that it is little benefit in upgrading from 802.11n 5 GHz to its successor 802.11ac in terms of performance and energy efficiency. We consider overall performance as well as the energy efficiency of 802.11n 2.4 GHz, 802.11n 5 GHz and 802.11ac protocols, all with 40MHz channels, to give a typical 802.11 office scenario. It is clear that 802.11ac can achieve slightly higher throughput compared to 802.11 for flows of large packets. However, the comparatively small benefits of 802.11ac may not justify the cost of buying and deploying new equipment for the upgrade. M. Abu-Tair, S. N. Bhatti. Energy Usage of UDP and DCCP over 802.11n. AINA 2014 - IEEE 28th Intl. Conf. Advanced Information Networking and Applications. Victoria, Canada. May 2014.
| PDF | .bib | 10.1109/AINA.2014.40 | abstract We show that the Datagram Congestion Control Protocol (DCCP) provides ~10% -- ~40% greater energy efficiency than the User Datagram Protocol (UDP) in a wireless LAN (WLAN) client. Our empirical evaluation uses a testbed comprised of consumer components and opensource software to measure typical performance that can be expected by a user, rather than highly-tuned performance which most users will not be able to configure. We focus our measurements on a scenario using IEEE 802.11n at 5GHz as energy efficiency will be particularly important to mobile and wireless users. We consider overall performance as well as the energy efficiency of the protocol usage to give a rounded comparison of UDP and DCCP. Overall, we see there would be great benefit in many applications using DCCP instead of UDP. M. Tauber, S. N. Bhatti. The Effect of the 802.11 Power Save Mechanism (PSM) on Energy Efficiency and Performance During System Activity. GreenCom 2012 - IEEE Intl. Conf. Green Computing and Communications. Besancon, France. Nov 2012.
| PDF | .bib | 10.1109/GreenCom.2012.81 | abstract 802.11 WLAN is a popular choice for wireless access on a range of ICT devices. A growing concern is the increased energy usage of ICT, for reasons of cost and environmental protection. The Power Save Mode (PSM) in 802.11 deactivates the wireless network interface during periods of inactivity. However, applications increasingly use push models, and so devices may be active much of the time. We have investigated the effectiveness of PSM, and considered its impact on performance when a device is active. Rather than concentrate on the NIC, we have taken a system-wide approach, to gauge the impact of the PSM from an application perspective. We experimentally evaluated performance at the packet level and system-wide power usage under various offered loads, controlled by packet size and data rate, on our 802.11n testbed. We have measured the system- wide power consumption corresponding to the individual traffic profiles and have derived application-specific effective energy-usage. We have found that in our scenarios, no significant benefit can be gained from using PSM. M. Tauber, S. N. Bhatti, Y. Yu. Towards Energy-Awareness In Managing Wireless LAN Applications. NOMS 2012 - IEEE Network Operations and Management Symp.. Maui, Hawaii, USA. Apr 2012.
| PDF | .bib | 10.1109/NOMS.2012.6211930 | abstract We have investigated the scope for enabling WLAN applications to manage the trade-off between performance and energy usage. We have conducted measurements of energy usage and performance in our 802.11n WLAN testbed, which operates in the 5 GHz ISM band. We have defined an effective energy usage envelope with respect to application-level packet transmission, and we demonstrate how performance as well as the effective energy usage envelope is effected by various configurations of IEEE 802.11n, including transmission power levels and channel width. Our findings show that the packet size and packet rate of the application flow have the greatest impact on application- level energy usage, compared to transmission power and channel width. As well as testing across a range of packet sizes and packet rates, we emulate a Skype flow, a YouTube flow and file transfers (HTTP over Internet and local server) to place our results in context. Based on our measurements we discuss approaches and potential improvements of management in effective energy usage for the tested applications. M. Tauber, S. N. Bhatti, Y. Yu. Application Level Energy and Performance Measurements in a Wireless LAN. GreenCom 2011 - IEEE/ACM Intl. Conf. Green Computing and Communications. Chengdu, Sichuan. Aug 2011.
| PDF | .bib | 10.1109/GreenCom.2011.26 | abstract We present an experimental evaluation of energy usage and performance in a wireless LAN cell based on a test bed using the 5 GHz ISM band for 802.11a and 802.11n. We have taken an application-level approach, by varying the packet size and transmission rate at the protocol level and evaluating energy usage across a range of application transmission rates using both large and small packet sizes. We have observed that both the application's transmission rate and the packet size have an impact on energy efficiency for transmission in our test bed. We also included in our experiments evaluation of the energy efficiency of emulations of YouTube and Skype flows, and a comparison with Ethernet transmissions.

Video

The growing use of video by users generates the largest volumes of traffic on the Internet. This work examines the kind of energy usage and performance we experience as users. This is through the definition and application of a new metric, the Energy Video Index (EnVI), that allows a combined assessment of user Quality of Experience (QoE) and energy usage for applications and end-systems that generate and consume video streams.

O. Ejembi, S. N. Bhatti. Go Green with EnVI: The Energy-Video Index. ISM 2015 - IEEE Intl. Symp. Multimedia. Miama, FL, USA. Dec 2015.
| PDF | .bib | 10.1109/ISM.2015.50 | abstract Video is the most prevalent traffic type on the Internet today. Significant research has been done on measuring user’s Quality of Experience (QoE) through different metrics. We take the position that energy use must be incorporated into quality metrics for digital video. We present our novel, energy- aware QoE metric for video, the Energy-Video Index (EnVI). We present our EnVI measurements from the playback of a diverse set of online videos. We observe that 4K-UHD (2160p) video can use ∼30% more energy on a client device compared to HD (720p), and up to ∼600% more network bandwidth than FHD (1080p), without significant improvement in objective QoE measurements. O. Ejembi, S. N. Bhatti. Client-side Energy Costs of Video Streaming. GreenCom 2015 - 11th IEEE Intl. Conf. Green Computing and Communications. Sydney, NSW, Australia. Dec 2015.
| PDF | .bib | 10.1109/DSDIS.2015.49 | abstract Through measurements on our testbed, we show how users of Netflix could make energy savings of up to 34% by adjusting video quality settings. We estimate the impacts of these quality settings on the energy consumption of client systems and the network. If users exercise choice in their video streaming habits, over 100 GWh of energy a year could be saved on a global scale. We discuss how providing energy usage information to users of digital video could enable them to make choices of video settings to reduce energy usage, and we estimate savings on associated electricity costs and carbon emissions. O. Ejembi, S. N. Bhatti. The Energy Cost of your Netflix Habit. e-Energy 2015 - 6th ACM Intl. Conf. Future Energy Systems. Bangalore, India. July 2015.
| PDF | .bib | 10.1145/2768510.2770951 | abstract Through measurements on our testbed, we show how users of Netflix could make energy savings of up to 34% by adjusting video quality settings. By using Netflix as a case study, we aim to assess the impact of energy usage in Video-on-demand (VoD) services. We estimate the potential impact of video quality settings on energy usage on a global scale. O. Ejembi, S. N. Bhatti. Towards Energy Benchmarking for Green Video. SustainIT 2015 - 4th IFIP Conference on Sustainable Internet and ICT for Sustainability. Madrid, Spain. April 2015.
| PDF | .bib | 10.1109/SustainIT.2015.7101375 | abstract Digital video is responsible for the largest proportion of traffic on the Internet today – upto ∼70\%. However, very little published research has examined the energy impact of this growing traffic type on a global scale (on client systems, servers and in the network). We summarise results and lessons learned from our measurement-based experiments on the energy use of digital video. By providing users with appropriate information and feedback, we could enable changes in user behaviour to save energy during use of digital video. We discuss the ongoing development of our benchmark tool which generates information on energy usage for users or other interested stakeholders. O. Ejembi, S. N. Bhatti. Help Save The Planet: Please Do Adjust Your Picture. MM 2014 - 22nd ACM Intl. Conf. Multimedia. Orlando, FL, USA. Nov 2014.
| .bib | 10.1145/2647868.2654897 | abstract Allowing digital video users to make choices of picture size and codec would significantly reduce energy usage, electricity costs and the carbon footprint of Internet users. Our empirical investigation shows a difference of up to a factor of 3 in energy usage for video decoding using different codecs at the same picture size and bitrate, on a desktop client system. With video traffic already responsible for the largest and fastest growing proportion of traffic on the Internet, a significant amount of energy, money and carbon output is due to video. We present a simple methodology and metrics that can be used to give an intuitive, quantitative and comparable assessment of the energy usage of video decoding. Providing energy usage information to users would empower them to make sensible choices. We demonstrate how small energy savings for individual client systems could give significant energy savings when considered at a global scale.

The behaviour of users, end-systems, and services

How can the use of end-systems and services provide information for users and applications so that they can change or adapt their energy-usage behaviour? An essentila part of the answer to this question is the ability to measure the energy usage of systems in use in a way that makes sense to users.

O. Ejembi, S. N. Bhatti. Client-side Energy Costs of Video Streaming. GreenCom 2015 - 11th IEEE Intl. Conf. Green Computing and Communications. Sydney, NSW, Australia. Dec 2015.
| PDF | .bib | 10.1109/DSDIS.2015.49 | abstract Through measurements on our testbed, we show how users of Netflix could make energy savings of up to 34% by adjusting video quality settings. We estimate the impacts of these quality settings on the energy consumption of client systems and the network. If users exercise choice in their video streaming habits, over 100 GWh of energy a year could be saved on a global scale. We discuss how providing energy usage information to users of digital video could enable them to make choices of video settings to reduce energy usage, and we estimate savings on associated electricity costs and carbon emissions. O. Ejembi, S. N. Bhatti. Help Save The Planet: Please Do Adjust Your Picture. MM 2014 - 22nd ACM Intl. Conf. Multimedia. Orlando, FL, USA. Nov 2014.
| .bib | 10.1145/2647868.2654897 | abstract Allowing digital video users to make choices of picture size and codec would significantly reduce energy usage, electricity costs and the carbon footprint of Internet users. Our empirical investigation shows a difference of up to a factor of 3 in energy usage for video decoding using different codecs at the same picture size and bitrate, on a desktop client system. With video traffic already responsible for the largest and fastest growing proportion of traffic on the Internet, a significant amount of energy, money and carbon output is due to video. We present a simple methodology and metrics that can be used to give an intuitive, quantitative and comparable assessment of the energy usage of video decoding. Providing energy usage information to users would empower them to make sensible choices. We demonstrate how small energy savings for individual client systems could give significant energy savings when considered at a global scale. Y. Yu, S. N. Bhatti. The cost of virtue: Reward as well as feedback are required to reduce user ICT power consumption. e-Energy 2014 - 5th ACM Intl. Conf. Future Energy Systems. Cambridge, UK. Jun 2014.
| .bib | 10.1145/2602044.2602063 | abstract We show that students in a school lab environment will change their behaviour to be more energy efficient, when appropriate incentives are in place, and when measurement-based, real-time feedback about their energy usage is provided. Rewards incentivise `non-green' users to be `green' as well as encouraging those users who already claim to be `green'. Measurement-based feedback improves user energy awareness and helps users to explore and adjust their use of computers to become `greener', but is not sufficient by itself. In our measurements, weekly mean group energy use as a whole reduced by up to 16\%; and weekly individual user energy consumption reduced by up to 56\% during active use. The findings are drawn from our longitudinal study that involved 83 Computer Science students; lasted 48 weeks across 2 academic years; monitored a total of 26778 hours of active computer use; collected approximately 3.2TB of raw data. M. Tauber, S. N. Bhatti. The Effect of the 802.11 Power Save Mechanism (PSM) on Energy Efficiency and Performance During System Activity. GreenCom 2012 - IEEE Intl. Conf. Green Computing and Communications. Besancon, France. Nov 2012.
| PDF | .bib | 10.1109/GreenCom.2012.81 | abstract 802.11 WLAN is a popular choice for wireless access on a range of ICT devices. A growing concern is the increased energy usage of ICT, for reasons of cost and environmental protection. The Power Save Mode (PSM) in 802.11 deactivates the wireless network interface during periods of inactivity. However, applications increasingly use push models, and so devices may be active much of the time. We have investigated the effectiveness of PSM, and considered its impact on performance when a device is active. Rather than concentrate on the NIC, we have taken a system-wide approach, to gauge the impact of the PSM from an application perspective. We experimentally evaluated performance at the packet level and system-wide power usage under various offered loads, controlled by packet size and data rate, on our 802.11n testbed. We have measured the system- wide power consumption corresponding to the individual traffic profiles and have derived application-specific effective energy-usage. We have found that in our scenarios, no significant benefit can be gained from using PSM. M. Tauber, S. N. Bhatti, Y. Yu. Towards Energy-Awareness In Managing Wireless LAN Applications. NOMS 2012 - IEEE Network Operations and Management Symp.. Maui, Hawaii, USA. Apr 2012.
| PDF | .bib | 10.1109/NOMS.2012.6211930 | abstract We have investigated the scope for enabling WLAN applications to manage the trade-off between performance and energy usage. We have conducted measurements of energy usage and performance in our 802.11n WLAN testbed, which operates in the 5 GHz ISM band. We have defined an effective energy usage envelope with respect to application-level packet transmission, and we demonstrate how performance as well as the effective energy usage envelope is effected by various configurations of IEEE 802.11n, including transmission power levels and channel width. Our findings show that the packet size and packet rate of the application flow have the greatest impact on application- level energy usage, compared to transmission power and channel width. As well as testing across a range of packet sizes and packet rates, we emulate a Skype flow, a YouTube flow and file transfers (HTTP over Internet and local server) to place our results in context. Based on our measurements we discuss approaches and potential improvements of management in effective energy usage for the tested applications. M. Tauber, S. N. Bhatti, Y. Yu. Application Level Energy and Performance Measurements in a Wireless LAN. GreenCom 2011 - IEEE/ACM Intl. Conf. Green Computing and Communications. Chengdu, Sichuan. Aug 2011.
| PDF | .bib | 10.1109/GreenCom.2011.26 | abstract We present an experimental evaluation of energy usage and performance in a wireless LAN cell based on a test bed using the 5 GHz ISM band for 802.11a and 802.11n. We have taken an application-level approach, by varying the packet size and transmission rate at the protocol level and evaluating energy usage across a range of application transmission rates using both large and small packet sizes. We have observed that both the application's transmission rate and the packet size have an impact on energy efficiency for transmission in our test bed. We also included in our experiments evaluation of the energy efficiency of emulations of YouTube and Skype flows, and a comparison with Ethernet transmissions. Y. Yu, S. N. Bhatti. Energy Measurement for the Cloud. ISPA 2010 - IEEE Intl. Symp. Parallel and Distributed Processing with Applications. Taipei, Taiwan. Sep 2010.
| PDF | .bib | 10.1109/ISPA.2010.29 | abstract One of the aims of cloud-based services is to offer cost savings through elastic service provision. This elasticity refers to use of resources by the customer and the provision of resources by the provider. An increasingly important resource to consider is energy (or power). As cloud services are intended to be ‘always on’ the energy costs of cloud service provision is already significant, and will continue to rise as global energy prices continue to rise and more users make use of cloud services. While providers can invest in energy-efficient hardware, how can we make users (customers) energy- aware, and incentivise users towards energy- efficient use of cloud systems? Clearly, being able to measure actual energy usage will allow the provision of suitable feedback to users, as well as enable new energy-aware systems metrics that allow systems management policies to become responsive to energy usage. We list the requirements and described a prototype for such an energy measurement system.