-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathawstream.tex
125 lines (103 loc) · 4.01 KB
/
awstream.tex
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
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
\documentclass[sigplan, 10pt, screen]{acmart}
\input{meta}
\begin{document}
\copyrightyear{2018}
\acmYear{2018}
\setcopyright{acmcopyright}
\acmConference[SIGCOMM '18]{ACM SIGCOMM 2018 Conference}{August 20--25, 2018}{Budapest, Hungary}
\acmBooktitle{SIGCOMM '18: ACM SIGCOMM 2018 Conference, August 20--25, 2018, Budapest, Hungary}
\acmPrice{15.00}
\acmDOI{10.1145/3230543.3230554}
\acmISBN{978-1-4503-5567-4/18/08}
\title{\sysname{}: Adaptive Wide-Area Streaming Analytics}
\author{Ben Zhang}
\affiliation{UC Berkeley}
\author{Xin Jin}
\affiliation{Johns Hopkins University}
\author{Sylvia Ratnasamy}
\affiliation{UC Berkeley}
\author{John Wawrzynek}
\affiliation{UC Berkeley}
\author{Edward A. Lee}
\affiliation{UC Berkeley}
\renewcommand{\shortauthors}{B. Zhang et al.}
\begin{abstract}
The emerging class of wide-area streaming analytics faces the challenge of
scarce and variable WAN bandwidth. Non-adaptive applications built with TCP or
UDP suffer from increased latency or degraded accuracy. State-of-the-art
approaches that adapt to network changes require developer writing sub-optimal
manual policies or are limited to application-specific optimizations.
We present \sysname{}, a stream processing system that simultaneously achieves
low latency and high accuracy in the wide area, requiring minimal developer
efforts. To realize this, \sysname{} uses three ideas: $(i)$ it integrates
application adaptation as a first-class programming abstraction in the stream
processing model; $(ii)$ with a combination of offline and online profiling, it
automatically learns an accurate profile that models accuracy and bandwidth
trade-off; and $(iii)$ at runtime, it carefully adjusts the application data
rate to match the available bandwidth while maximizing the achievable
accuracy. We evaluate \sysname{} with three real-world applications: augmented
reality, pedestrian detection, and monitoring log analysis. Our experiments show
that \sysname{} achieves sub-second latency with only nominal accuracy drop
(2-6\%).
\end{abstract}
\begin{CCSXML}
<ccs2012>
<concept>
<concept_id>10003033.10003039.10003051</concept_id>
<concept_desc>Networks~Application layer protocols</concept_desc>
<concept_significance>500</concept_significance>
</concept>
<concept>
<concept_id>10003033.10003039.10003056</concept_id>
<concept_desc>Networks~Cross-layer protocols</concept_desc>
<concept_significance>500</concept_significance>
</concept>
<concept>
<concept_id>10003033.10003079.10003080</concept_id>
<concept_desc>Networks~Network performance modeling</concept_desc>
<concept_significance>500</concept_significance>
</concept>
<concept>
<concept_id>10003033.10003083.10003084.10003088</concept_id>
<concept_desc>Networks~Wide area networks</concept_desc>
<concept_significance>500</concept_significance>
</concept>
<concept>
<concept_id>10003033.10003083.10003094</concept_id>
<concept_desc>Networks~Network dynamics</concept_desc>
<concept_significance>500</concept_significance>
</concept>
<concept>
<concept_id>10003033.10003106.10010924</concept_id>
<concept_desc>Networks~Public Internet</concept_desc>
<concept_significance>500</concept_significance>
</concept>
</ccs2012>
\end{CCSXML}
\ccsdesc[500]{Networks~Application layer protocols}
\ccsdesc[500]{Networks~Cross-layer protocols}
\ccsdesc[500]{Networks~Network performance modeling}
\ccsdesc[500]{Networks~Wide area networks}
\ccsdesc[500]{Networks~Network dynamics}
\ccsdesc[500]{Networks~Public Internet}
\keywords{Wide Area Network; Adaptation; Learning; Profiling}
\maketitle
\input{sections/intro}
\input{sections/motivation}
\input{sections/system}
\input{sections/impl}
\input{sections/evaluation}
\input{sections/discussion}
\input{sections/related}
\input{sections/conclusion}
\input{sections/acknowledgment}
% \input{sections/appendix.tex}
\balance
\bibliographystyle{ACM-Reference-Format}
% \def \bibfont {\normalsize}
\bibliography{awstream}
\end{document}
%%% Local Variables:
%%% mode: latex
%%% TeX-master: t
%%% End: