utility.tex
3.4 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
\section{Computing Energy Efficiency}
\label{sec-utility}
Based on the results from the previous section, we can formulate design
requirements for an energy efficiency metric to apply to smartphone apps.
First, it must scale with usage, respecting the differences in baseline
consumption between users identified in Section~\ref{subsec-uservariation}
and the temporal variation of apps identified in
Section~\ref{subsec-timevariation}. Second, it must try to avoid targeting
top apps, even if they tend to consume a great deal of energy as described in
Section~\ref{subsec-consumption}, as these may not be apps that users are
willing to uninstall. Finally, we use the analysis of background energy
consumption in Section~\ref{subsec-background} as a hint about where to
start, given that background energy consumption should match foreground usage
in most cases.
In the section we walk through several ways of characterizing app energy
consumption: via the total amount, by consumption rate, and scaled against
foreground energy consumption and a new content-delivery metric we design
that incorporates use of both the display and the audio device. In each case
we examine the app consumption data generated by our usage monitoring study
and use each metric to shed light on app energy consumption.
\subsection{Total Consumption}
\input{./figures/tables/tableTOTAL.tex}
Clearly, ranking apps by total energy consumption over the entire study says
much more about app popularity than it does about anything else.
Table~\ref{table-total} shows the top and bottom energy-consuming apps over
the entire study. As expected, popular apps such as the Android Browser,
Facebook, and the Android Phone compunent consume the most energy, while the
list of low consumers is dominated by apps with few installs. This table does
serve, however, to identify the popular apps in use by \PhoneLab{}
participants.
\subsection{Consumption Rate}
\input{./figures/tables/tableRATE.tex}
Computing the rate at which apps consume energy by scaling their total energy
usage against the total time they were running, either in the background or
foreground, reveals more information, as shown in Table~\ref{table-rate}, The
results identify Facebook Messenger, Google+, and the Super-Bright LED
Flashlight as apps that rapidly-consume energy, while the Bank of America and
Weather Channel apps consume energy slowly. Differences between apps in
similar categories may begin to identify apps with problematic energy
consumption, such as contrasting the high energy usage of Facebook Messenger
with the low usage of WhatsApp, Twitter, Android Messaging, and even Skype.
\subsection{Foreground Energy Efficiency}
\input{./figures/tables/tableFOREGROUND.tex}
Consumption rate alone, however, is insufficient to answer important
questions about how efficient smartphone apps are. Pandora, for example, may
consume a great deal of energy either because it is poorly written, or
because it is delivering a great deal of content. Given the observations
about background usage presented earlier, we were interested in using an apps
foreground time as a utility metric to compute energy efficiency. In this
conceptual framework, smartphone apps deliver utility through screen time
with users, and should consume energy in proportion to the amount of time
users spend actively interacting with them.
\subsection{Content Energy Efficiency}
\input{./figures/tables/tableCONTENT.tex}
\subsection{Discussion}