An Analysis of Power Consumption in a Smartphone

By A. Carroll, Gernot Heiser. 2010

1,673 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Federated Multi-Task Learning (2017, methodology): “Communication can also exacerbate poor performance, as it can be slower than computation by many orders of magnitude in typical cellular or wireless networks [49, 19, 46, 8, 36].”
AutoScale: Energy Efficiency Optimization for Stochastic Edge Inference Using Reinforcement Learning (2020, methodology): “We measure the system-wide power consumption of the smartphones using an external Monsoon Power Meter [80]; prior works used the similar practice [8], [11], [88].”
Climate change and COP26: Are digital technologies and information management part of the problem or the solution? An editorial reflection and call to action (2022, background): “Future projections relating to Smart Cities, 5 G and the Internet of Things give rise to additional concerns over energy demand Carroll and Heiser (2010).”
EnerJ: approximate data types for safe and general low-power computation (2011, background): “In a smartphone, CPU and memory account for about 20% and the radio typically close to 50% of the overall power [6].”
AppScope: Application Energy Metering Framework for Android Smartphone Using Kernel Activity Monitoring (2012, methodology): “Usage estimation for hardware components has been previously completed using hardware performance counter (HPC) [5, 10-13], software performance counter (SPC) such as the Linux procfs/sysfs [7, 14-16], or BatteryStats, which is provided by Android [8, 17].”
A Survey on Mobile Edge Computing: The Communication Perspective (2017, background): “, the random access memory (RAM) and flash memory, also contribute to the computation latency and energy consumption [80], while detailed discussions are beyond the scope of this survey.”
Google Workloads for Consumer Devices: Mitigating Data Movement Bottlenecks (2018, background): “Prior works [15, 57, 68, 113, 121] study the general performance and energy profile of mobile applications.”
Dynamic voltage and frequency scaling: the laws of diminishing returns (2010, methodology): “We have also only considered the effectiveness of DVFS on server-class platforms, while DVFS may still be effective on other platform, such as smart-phones [3].”
Where is the energy spent inside my app?: fine grained energy accounting on smartphones with Eprof (2012, background): “Heiser, Ananalysisofpowerconsumption in a smartphone, in Proc. of USENIX ATC,2010.”
Flikker: saving DRAM refresh-power through critical data partitioning (2011, methodology): “Based on previous study of DRAM power as a fraction of total power consumption, this 20-25% corresponds to 1% of total power savings in a smartphone [6].”
A study of mobile device utilization (2015, background): “The processor takes up a substantial portion of power consumption in mobile devices [16].”
AutoScale: Optimizing Energy Efficiency of End-to-End Edge Inference under Stochastic Variance (2020, methodology): “the system-wide power consumption of the smartphones using an external Monsoon Power Meter [68] – similar practice is used in a number of prior works [6, 9, 75].”
Sentinel: occupancy based HVAC actuation using existing WiFi infrastructure within commercial buildings (2013, background): “Prior works on WiFi and cellular radio power measurements [16, 19, 51] indicate that WiFi sleep power is about 2x the sleep power of cellular technologies such as 3G, the data transmission in WiFi is about an order of magnitude more efficient than cellular, the energy spent by WiFi radio to scan and”
Measuring energy consumption for short code paths using RAPL (2012, background): “This can be leveraged by more fine-grained instrumentation of single processing units [2].”
Quantifying the energy cost of data movement for emerging smart phone workloads on mobile platforms (2014, methodology): “expect a significant portion of the Other energy consumption to come from the smart phone display, which has been shown as one of the most power-hungry components in modern smart phones [1], [2].”
Anomaly Detection in Vertically Partitioned Data by Distributed Core Vector Machines (2013, background): “Communication is one of the most expensive operations in wireless networks of battery-powered sensors [3] and mobile devices [6].”