blue / papers.hotmobile2015-numerator · Commits

Commit 01816115f50310a452421e0e89bb6751a66044b3

Authored by Geoffrey Challen
1 parent ebefbc89

Working.

Inline Side-by-side
Showing 2 changed files with 25 additions and 14 deletions
figures/tables/tableALL.tex
  View file @0181611
... ... @@ -52,6 +52,7 @@
52 52 \num{10} & \texttt{Android News and Weather} & 0.254 \\
53 53 \bottomrule
54 54 \bottomrule
  55 +\num{10} & & \\
55 56 \num{9} & \texttt{WhatsApp Messenger} & 0.095 \\
56 57 \num{8} & \texttt{Twitter} & 0.078 \\
57 58 \num{7} & \texttt{Yahoo Mail} & 0.077 \\
... ... @@ -85,6 +86,7 @@
85 86 \num{7} & \texttt{Android Calculator} & 9.189 \\
86 87 \num{8} & \texttt{Twitter} & 8.645 \\
87 88 \num{9} & \texttt{Chrome Browser} & 8.524 \\
  89 +\num{10} & & \\
88 90 \bottomrule
89 91 \num{10} & \texttt{Yahoo Mail} & 3.287 \\
90 92 \num{9} & \texttt{ESPN SportsCenter} & 3.184 \\
... ... @@ -101,6 +103,8 @@
101 103  
102 104 \caption{\small \textbf{Apps sorted by foreground energy efficiency.}}
103 105  
  106 +\label{table-foreground}
  107 +
104 108 \end{subtable}%
105 109 \begin{subtable}[t]{0.5\textwidth}
106 110 {\small
... ... @@ -118,6 +122,7 @@
118 122 \num{7} & \texttt{Twitter} & 5610.394 \\
119 123 \num{8} & \texttt{Android Clock} & 5085.873 \\
120 124 \num{9} & \texttt{Yahoo Mail} & 5083.615 \\
  125 +\num{10} & & \\
121 126 \bottomrule
122 127 \num{10} & \texttt{NFL Mobile} & 1275.985 \\
123 128 \num{9} & \texttt{UB Parking} & 1071.529 \\
... ... @@ -132,6 +137,8 @@
132 137 \end{tabularx}
133 138 }
134 139  
  140 +\label{table-content}
  141 +
135 142 \caption{\small \textbf{Apps sorted by content energy efficiency.}}
136 143 \end{subtable}
137 144  
... ...
results.tex
  View file @0181611
... ... @@ -39,7 +39,7 @@ it through a survey completed by 47~experiment participants. Unfortunately,
39 39 our results are inconclusive and open to several possible interpretations
40 40 which we conclude by discussing.
41 41  
42   -\subsection{Total Consumption}
  42 +\subsection{Total Energy}
43 43  
44 44 \input{./figures/tables/tableALL.tex}
45 45  
... ... @@ -52,29 +52,33 @@ list of low consumers is dominated by apps with few installs. This table does
52 52 serve, however, to identify the popular apps in use by \PhoneLab{}
53 53 participants, and as a point of comparison for the remainder of our results.
54 54  
55   -\subsection{Consumption Rate}
  55 +\subsection{Power}
56 56  
57   -Computing the rate at which apps consume energy by scaling their total energy
58   -usage against the total time they were running, either in the background or
  57 +Computing each app's power consumption by scaling their total energy usage
  58 +against the total time they were running, either in the background or
59 59 foreground, reveals more information, as shown in Table~\ref{table-rate}. Our
60 60 results identify Facebook Messenger, Google+, and the Super-Bright LED
61 61 Flashlight as apps that rapidly-consume energy, while the Bank of America and
62 62 Weather Channel apps consume energy slowly. Differences between apps in
63 63 similar categories may begin to identify apps with problematic energy
64 64 consumption, such as contrasting the high energy usage of Facebook Messenger
65   -with other messaging clients such as WhatsApp, Twitter, and Android Messaging.
  65 +with other messaging clients such as WhatsApp, Twitter, and Android
  66 +Messaging.
66 67  
67 68 \subsection{Foreground Energy Efficiency}
68 69  
69   -Consumption rate alone, however, is insufficient to answer important
70   -questions about how efficient smartphone apps are. Pandora, for example, may
71   -consume a great deal of energy either because it is poorly written, or
72   -because it is delivering a great deal of content. Given the observations
73   -about background usage presented earlier, we were interested in using an apps
74   -foreground time as a utility metric to compute energy efficiency. In this
75   -conceptual framework, smartphone apps deliver utility through screen time
76   -with users, and should consume energy in proportion to the amount of time
77   -users spend actively interacting with them.
  70 +Isolating the foreground component of execution time provides a better
  71 +measure of value, since it ignores the time that users spend ignoring apps.
  72 +Table~\ref{table-foreground} shows a measure of energy efficiency computed by
  73 +utilizing foreground time alone as our value measure. Some surprising changes
  74 +from the power results can be seen. Some apps have remaining in their former
  75 +categories: Bank of America, which was identified as a low-power app, is also
  76 +a highly-efficient app when using foreground time as the value measure; and
  77 +Facebook Messenger, which was identified as a high-power app, is also marked
  78 +as inefficient. Other apps, however, have switched categories. ESPN
  79 +Sportscenter and Yahoo Mail do not consume much power, but also don't spend
  80 +much time in the foreground; interestingly, none of the high-power apps
  81 +looked better when their foreground usage was considered.
78 82  
79 83 \subsection{Content Energy Efficiency}
80 84  
... ...