forked from FPBench/fpbench.org
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathbenchmarks.html
220 lines (199 loc) · 6.86 KB
/
benchmarks.html
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
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
<!doctype html>
<html>
<head>
<meta charset="utf-8"/>
<title>FPBench Benchmarks</title>
<link rel="stylesheet" type="text/css" href="fpbench.css">
</head>
<body>
<header>
<a href='.' style='color: black; text-decoration: none;'>
<img src='img/logo.png' height='150' alt='FPBench Logo' />
<h1>FPBench Benchmarks</h1>
<p>Standard benchmarks for floating-point research</p>
</a>
</header>
<p><a href=".">FPBench</a> includes a suite of benchmarks, drawn
from published floating-point research.</p>
<figure class="columns">
<table class="counts">
<thead>
<th colspan="2">Benchmark sources</th>
</thead>
<tr><td>Rosa</td><td>37</td></tr>
<tr><td>Herbie</td><td>28</td></tr>
<tr><td>Salsa</td><td>25</td></tr>
<tr><td>FPTaylor</td><td>21</td></tr>
</table>
<table class="counts">
<thead>
<th colspan="2">Features used</th>
</thead>
<tr><td>Arithmetic</td><td>111</td></tr>
<tr><td>Temporaries</td><td>57</td></tr>
<tr><td>Comparison</td><td>33</td></tr>
<tr><td>Loops</td><td>28</td></tr>
<tr><td>Exponents</td><td>16</td></tr>
<tr><td>Trigonometry</td><td>15</td></tr>
<tr><td>Conditionals</td><td>10</td></tr>
</table>
<table class="counts">
<thead>
<th colspan="2">Domains</th>
</thead>
<tr><td>Textbooks</td><td>28</td></tr>
<tr><td>Mathematics</td><td>24</td></tr>
<tr><td>Controls</td><td>10</td></tr>
<tr><td>Science</td><td>10</td></tr>
<tr><td>(unknown)</td><td>39</td></tr>
</table>
</figure>
<p>These benchmarks have been contributed by multiple research
groups. These contributions are listed here in chronological order.
Each listed paper has all its evaluation examples incorporated into
FPBench.</p>
<p>For each benchmark, FPBench records its name (preferably as used
in a paper), descriptive information such as citations, and finally
semantic metadata such as the precision used and preconditions.</p>
<h2>The CPU group at the University of Utah</h2>
<figure class="columns">
<img class="logo" src="img/uu-cpu.png" />
<table class="counts">
<thead>
<th colspan="2">Features used</th>
</thead>
<tr><td>Arithmetic</td><td>21</td></tr>
<tr><td>Temporaries</td><td>9</td></tr>
<tr><td>Exponents</td><td>3</td></tr>
<tr><td>Trigonometry</td><td>2</td></tr>
</table>
<table class="counts">
<thead>
<th colspan="2">Domains</th>
</thead>
<tr><td>(unknown)</td><td>21</td></tr>
</table>
</figure>
<p>
The <a href="http://parallel.utah.edu/">Center for Parallel
Computing at Utah</a> group's FPTaylor project contributed 21
benchmarks, many short textbook examples and mathematical
computations.
</p>
<ul class="paper-list">
<li>
Rigorous Estimation of Floating-Point Round-off Errors with Symbolic Taylor Expansions<br/>
Alexey Solovyev, Charlie Jacobsen, Zvonimir Rakamaric, Ganesh Gopalakrishnan, <i>FM'15</i>
</li>
</ul>
<h2>The AVA group at the Max Plank Institute for Software Systems</h2>
<figure class="columns">
<img class="logo" src="img/mpi-ava.png" />
<table class="counts">
<thead>
<th colspan="2">Features used</th>
</thead>
<tr><td>Arithmetic</td><td>37</td></tr>
<tr><td>Temporaries</td><td>23</td></tr>
<tr><td>Comparison</td><td>8</td></tr>
<tr><td>Conditionals</td><td>5</td></tr>
<tr><td>Loops</td><td>3</td></tr>
<tr><td>Exponents</td><td>1</td></tr>
<tr><td>Trigonometry</td><td>1</td></tr>
</table>
<table class="counts">
<thead>
<th colspan="2">Domains</th>
</thead>
<tr><td>Science</td><td>10</td></tr>
<tr><td>Mathematics</td><td>5</td></tr>
<tr><td>Controls</td><td>4</td></tr>
<tr><td>(unknown)</td><td>18</td></tr>
</table>
</figure>
<p>
The Automated Verification and Approximation group's Rosa project
contributed 37 benchmarks, mostly drawn from scientific
computations and control theory.
</p>
<ul class="paper-list">
<li>
Sound Compilation of the Reals<br/>
Eva Darulova, Viktor Kuncak, <i>POPL'14</i>
</li>
</ul>
<h2>The LAMPS group at the Université de Perpignan Via Domitia</h2>
<figure class="columns">
<img class="logo" src="img/upd-lamps.png" />
<table class="counts">
<thead>
<th colspan="2">Features used</th>
</thead>
<tr><td>Arithmetic</td><td>25</td></tr>
<tr><td>Temporaries</td><td>25</td></tr>
<tr><td>Loops</td><td>25</td></tr>
<tr><td>Comparison</td><td>25</td></tr>
<tr><td>Conditionals</td><td>5</td></tr>
<tr><td>Trigonometry</td><td>4</td></tr>
</table>
<table class="counts">
<thead>
<th colspan="2">Domains</th>
</thead>
<tr><td>Mathematics</td><td>19</td></tr>
<tr><td>Controls</td><td>6</td></tr>
</table>
</figure>
<p>
The <a href="http://lamps.univ-perp.fr/LAMPS/Accueil.html">Laboratoire
de Mathématiques et de Physique</a> group's Salsa project
contributed 25 benchmarks, all programs with loops drawn from
mathematical algorithms and control theory.
</p>
<ul class="paper-list">
<li>
Intra-procedural Optimization of the Numerical Accuracy of Programs<br/>
Nasrine Damouche, Matthieu Martel, Alexandre Chapoutot, <i>FMICS'15</i>
</li>
<li>
Transformation of a PID Controller for Numerical Accuracy<br/>
Nasrine Damouche, Matthieu Martel, Alexandre Chapoutot, <i>NSV'15</i>
</li>
<li>
Optimizing the Accuracy of a Rocket Trajectory Simulation by Program Transformation<br/>
Nasrine Damouche, Matthieu Martel, Alexandre Chapoutot, <i>CF'15</i>
</li>
</ul>
<h2>The PLSE group at the University of Washington</h2>
<figure class="columns">
<img class="logo" src="img/uw-plse.jpg" />
<table class="counts">
<thead>
<th colspan="2">Features used</th>
</thead>
<tr><td>Arithmetic</td><td>28</td></tr>
<tr><td>Exponents</td><td>12</td></tr>
<tr><td>Trigonometry</td><td>8</td></tr>
</table>
<table class="counts">
<thead>
<th colspan="2">Domains</th>
</thead>
<tr><td>Textbooks</td><td>28</td></tr>
</table>
</figure>
<p>
The <a href="https://uwplse.org">Programming Languages and
Software Engineering</a>
group's <a href="https://herbie.uwplse.org">Herbie</a> project
contributed 28 benchmarks, mostly drawn from Richard
Hamming's <i>Numerical Methods for Scientists and Engineers</i>.
</p>
<ul class="paper-list">
<li>
<a href="http://herbie.uwplse.org/pldi15.html">Automatically Improving Accuracy for Floating-point Expressions</a><br/>
Pavel Panchekha, Alex Sanchez-Stern, James R. Wilcox, Zachary Tatlock, <i>PLDI'15</i>
</li>
</ul>
</body>
</html>