-
Notifications
You must be signed in to change notification settings - Fork 1
Files
/
benchmark_core_distance_blocking.csv
Latest commit
433 lines (433 loc) · 14 KB
/
benchmark_core_distance_blocking.csv
1 | NAME | NUM_POINTS | DIM | CYC |
---|---|---|---|---|
2 | Baseline | 1024 | 2 | 12049 |
3 | 2-fold Unrolled | 1024 | 2 | 12004 |
4 | 4-fold Unrolled | 1024 | 2 | 13451 |
5 | AVX Vectorized | 1024 | 2 | 12825 |
6 | AVX Vectorized FMA | 1024 | 2 | 12772 |
7 | Symmetry+Blocking | 1024 | 2 | 14739 |
8 | Blocking Optimized | 1024 | 2 | 10743 |
9 | Blocking Optimized & Adaptive | 1024 | 2 | 10938 |
10 | Baseline | 1024 | 4 | 13714 |
11 | 2-fold Unrolled | 1024 | 4 | 13482 |
12 | 4-fold Unrolled | 1024 | 4 | 13898 |
13 | AVX Vectorized | 1024 | 4 | 13230 |
14 | AVX Vectorized FMA | 1024 | 4 | 13861 |
15 | Symmetry+Blocking | 1024 | 4 | 15036 |
16 | Blocking Optimized | 1024 | 4 | 10080 |
17 | Blocking Optimized & Adaptive | 1024 | 4 | 9756 |
18 | Baseline | 1024 | 8 | 17600 |
19 | 2-fold Unrolled | 1024 | 8 | 15516 |
20 | 4-fold Unrolled | 1024 | 8 | 15464 |
21 | AVX Vectorized | 1024 | 8 | 14208 |
22 | AVX Vectorized FMA | 1024 | 8 | 15485 |
23 | Symmetry+Blocking | 1024 | 8 | 15579 |
24 | Blocking Optimized | 1024 | 8 | 10079 |
25 | Blocking Optimized & Adaptive | 1024 | 8 | 11035 |
26 | Baseline | 1024 | 16 | 26582 |
27 | 2-fold Unrolled | 1024 | 16 | 23022 |
28 | 4-fold Unrolled | 1024 | 16 | 21255 |
29 | AVX Vectorized | 1024 | 16 | 17000 |
30 | AVX Vectorized FMA | 1024 | 16 | 15924 |
31 | Symmetry+Blocking | 1024 | 16 | 17018 |
32 | Blocking Optimized | 1024 | 16 | 10818 |
33 | Blocking Optimized & Adaptive | 1024 | 16 | 10375 |
34 | Baseline | 1024 | 32 | 51000 |
35 | 2-fold Unrolled | 1024 | 32 | 36317 |
36 | 4-fold Unrolled | 1024 | 32 | 30089 |
37 | AVX Vectorized | 1024 | 32 | 19784 |
38 | AVX Vectorized FMA | 1024 | 32 | 21254 |
39 | Symmetry+Blocking | 1024 | 32 | 30504 |
40 | Blocking Optimized | 1024 | 32 | 13095 |
41 | Blocking Optimized & Adaptive | 1024 | 32 | 12900 |
42 | Baseline | 1024 | 64 | 91330 |
43 | 2-fold Unrolled | 1024 | 64 | 72088 |
44 | 4-fold Unrolled | 1024 | 64 | 52088 |
45 | AVX Vectorized | 1024 | 64 | 27173 |
46 | AVX Vectorized FMA | 1024 | 64 | 26985 |
47 | Symmetry+Blocking | 1024 | 64 | 26868 |
48 | Blocking Optimized | 1024 | 64 | 16858 |
49 | Blocking Optimized & Adaptive | 1024 | 64 | 18815 |
50 | Baseline | 1024 | 128 | 198668 |
51 | 2-fold Unrolled | 1024 | 128 | 157402 |
52 | 4-fold Unrolled | 1024 | 128 | 102747 |
53 | AVX Vectorized | 1024 | 128 | 44655 |
54 | AVX Vectorized FMA | 1024 | 128 | 42209 |
55 | Symmetry+Blocking | 1024 | 128 | 37962 |
56 | Blocking Optimized | 1024 | 128 | 22079 |
57 | Blocking Optimized & Adaptive | 1024 | 128 | 22608 |
58 | Baseline | 1024 | 256 | 409393 |
59 | 2-fold Unrolled | 1024 | 256 | 333043 |
60 | 4-fold Unrolled | 1024 | 256 | 191162 |
61 | AVX Vectorized | 1024 | 256 | 74603 |
62 | AVX Vectorized FMA | 1024 | 256 | 73983 |
63 | Symmetry+Blocking | 1024 | 256 | 72578 |
64 | Blocking Optimized | 1024 | 256 | 39708 |
65 | Blocking Optimized & Adaptive | 1024 | 256 | 44236 |
66 | Baseline | 1024 | 512 | 852447 |
67 | 2-fold Unrolled | 1024 | 512 | 669720 |
68 | 4-fold Unrolled | 1024 | 512 | 377537 |
69 | AVX Vectorized | 1024 | 512 | 129415 |
70 | AVX Vectorized FMA | 1024 | 512 | 131889 |
71 | Symmetry+Blocking | 1024 | 512 | 118854 |
72 | Blocking Optimized | 1024 | 512 | 129110 |
73 | Blocking Optimized & Adaptive | 1024 | 512 | 67724 |
74 | Baseline | 2048 | 2 | 28082 |
75 | 2-fold Unrolled | 2048 | 2 | 28060 |
76 | 4-fold Unrolled | 2048 | 2 | 31114 |
77 | AVX Vectorized | 2048 | 2 | 29367 |
78 | AVX Vectorized FMA | 2048 | 2 | 30250 |
79 | Symmetry+Blocking | 2048 | 2 | 34908 |
80 | Blocking Optimized | 2048 | 2 | 25240 |
81 | Blocking Optimized & Adaptive | 2048 | 2 | 25308 |
82 | Baseline | 2048 | 4 | 31917 |
83 | 2-fold Unrolled | 2048 | 4 | 31030 |
84 | 4-fold Unrolled | 2048 | 4 | 32504 |
85 | AVX Vectorized | 2048 | 4 | 31613 |
86 | AVX Vectorized FMA | 2048 | 4 | 31660 |
87 | Symmetry+Blocking | 2048 | 4 | 32764 |
88 | Blocking Optimized | 2048 | 4 | 24898 |
89 | Blocking Optimized & Adaptive | 2048 | 4 | 24091 |
90 | Baseline | 2048 | 8 | 40536 |
91 | 2-fold Unrolled | 2048 | 8 | 36673 |
92 | 4-fold Unrolled | 2048 | 8 | 36560 |
93 | AVX Vectorized | 2048 | 8 | 33303 |
94 | AVX Vectorized FMA | 2048 | 8 | 33561 |
95 | Symmetry+Blocking | 2048 | 8 | 37212 |
96 | Blocking Optimized | 2048 | 8 | 26084 |
97 | Blocking Optimized & Adaptive | 2048 | 8 | 25881 |
98 | Baseline | 2048 | 16 | 60712 |
99 | 2-fold Unrolled | 2048 | 16 | 50327 |
100 | 4-fold Unrolled | 2048 | 16 | 46896 |
101 | AVX Vectorized | 2048 | 16 | 38319 |
102 | AVX Vectorized FMA | 2048 | 16 | 38027 |
103 | Symmetry+Blocking | 2048 | 16 | 39907 |
104 | Blocking Optimized | 2048 | 16 | 27481 |
105 | Blocking Optimized & Adaptive | 2048 | 16 | 27909 |
106 | Baseline | 2048 | 32 | 102060 |
107 | 2-fold Unrolled | 2048 | 32 | 66610 |
108 | 4-fold Unrolled | 2048 | 32 | 59839 |
109 | AVX Vectorized | 2048 | 32 | 39246 |
110 | AVX Vectorized FMA | 2048 | 32 | 38876 |
111 | Symmetry+Blocking | 2048 | 32 | 39297 |
112 | Blocking Optimized | 2048 | 32 | 28199 |
113 | Blocking Optimized & Adaptive | 2048 | 32 | 30541 |
114 | Baseline | 2048 | 64 | 185973 |
115 | 2-fold Unrolled | 2048 | 64 | 151946 |
116 | 4-fold Unrolled | 2048 | 64 | 114913 |
117 | AVX Vectorized | 2048 | 64 | 56209 |
118 | AVX Vectorized FMA | 2048 | 64 | 55549 |
119 | Symmetry+Blocking | 2048 | 64 | 57967 |
120 | Blocking Optimized | 2048 | 64 | 39770 |
121 | Blocking Optimized & Adaptive | 2048 | 64 | 38758 |
122 | Baseline | 2048 | 128 | 368498 |
123 | 2-fold Unrolled | 2048 | 128 | 296526 |
124 | 4-fold Unrolled | 2048 | 128 | 186099 |
125 | AVX Vectorized | 2048 | 128 | 87124 |
126 | AVX Vectorized FMA | 2048 | 128 | 90497 |
127 | Symmetry+Blocking | 2048 | 128 | 85168 |
128 | Blocking Optimized | 2048 | 128 | 57595 |
129 | Blocking Optimized & Adaptive | 2048 | 128 | 56790 |
130 | Baseline | 2048 | 256 | 827277 |
131 | 2-fold Unrolled | 2048 | 256 | 655535 |
132 | 4-fold Unrolled | 2048 | 256 | 372272 |
133 | AVX Vectorized | 2048 | 256 | 152398 |
134 | AVX Vectorized FMA | 2048 | 256 | 142077 |
135 | Symmetry+Blocking | 2048 | 256 | 133441 |
136 | Blocking Optimized | 2048 | 256 | 85043 |
137 | Blocking Optimized & Adaptive | 2048 | 256 | 84099 |
138 | Baseline | 2048 | 512 | 1678916 |
139 | 2-fold Unrolled | 2048 | 512 | 1362574 |
140 | 4-fold Unrolled | 2048 | 512 | 734257 |
141 | AVX Vectorized | 2048 | 512 | 256944 |
142 | AVX Vectorized FMA | 2048 | 512 | 263949 |
143 | Symmetry+Blocking | 2048 | 512 | 242794 |
144 | Blocking Optimized | 2048 | 512 | 254418 |
145 | Blocking Optimized & Adaptive | 2048 | 512 | 171391 |
146 | Baseline | 4096 | 2 | 72060 |
147 | 2-fold Unrolled | 4096 | 2 | 65849 |
148 | 4-fold Unrolled | 4096 | 2 | 68374 |
149 | AVX Vectorized | 4096 | 2 | 69328 |
150 | AVX Vectorized FMA | 4096 | 2 | 69013 |
151 | Symmetry+Blocking | 4096 | 2 | 76164 |
152 | Blocking Optimized | 4096 | 2 | 60049 |
153 | Blocking Optimized & Adaptive | 4096 | 2 | 56894 |
154 | Baseline | 4096 | 4 | 71737 |
155 | 2-fold Unrolled | 4096 | 4 | 71761 |
156 | 4-fold Unrolled | 4096 | 4 | 72690 |
157 | AVX Vectorized | 4096 | 4 | 71166 |
158 | AVX Vectorized FMA | 4096 | 4 | 70437 |
159 | Symmetry+Blocking | 4096 | 4 | 74221 |
160 | Blocking Optimized | 4096 | 4 | 58620 |
161 | Blocking Optimized & Adaptive | 4096 | 4 | 56735 |
162 | Baseline | 4096 | 8 | 88826 |
163 | 2-fold Unrolled | 4096 | 8 | 82725 |
164 | 4-fold Unrolled | 4096 | 8 | 82547 |
165 | AVX Vectorized | 4096 | 8 | 74979 |
166 | AVX Vectorized FMA | 4096 | 8 | 75709 |
167 | Symmetry+Blocking | 4096 | 8 | 81969 |
168 | Blocking Optimized | 4096 | 8 | 59795 |
169 | Blocking Optimized & Adaptive | 4096 | 8 | 59907 |
170 | Baseline | 4096 | 16 | 127409 |
171 | 2-fold Unrolled | 4096 | 16 | 107430 |
172 | 4-fold Unrolled | 4096 | 16 | 100919 |
173 | AVX Vectorized | 4096 | 16 | 83819 |
174 | AVX Vectorized FMA | 4096 | 16 | 84605 |
175 | Symmetry+Blocking | 4096 | 16 | 87585 |
176 | Blocking Optimized | 4096 | 16 | 63758 |
177 | Blocking Optimized & Adaptive | 4096 | 16 | 62781 |
178 | Baseline | 4096 | 32 | 221889 |
179 | 2-fold Unrolled | 4096 | 32 | 165227 |
180 | 4-fold Unrolled | 4096 | 32 | 138094 |
181 | AVX Vectorized | 4096 | 32 | 96462 |
182 | AVX Vectorized FMA | 4096 | 32 | 96888 |
183 | Symmetry+Blocking | 4096 | 32 | 98436 |
184 | Blocking Optimized | 4096 | 32 | 72162 |
185 | Blocking Optimized & Adaptive | 4096 | 32 | 76378 |
186 | Baseline | 4096 | 64 | 376133 |
187 | 2-fold Unrolled | 4096 | 64 | 310161 |
188 | 4-fold Unrolled | 4096 | 64 | 232174 |
189 | AVX Vectorized | 4096 | 64 | 126923 |
190 | AVX Vectorized FMA | 4096 | 64 | 126462 |
191 | Symmetry+Blocking | 4096 | 64 | 130303 |
192 | Blocking Optimized | 4096 | 64 | 90509 |
193 | Blocking Optimized & Adaptive | 4096 | 64 | 91133 |
194 | Baseline | 4096 | 128 | 794003 |
195 | 2-fold Unrolled | 4096 | 128 | 599658 |
196 | 4-fold Unrolled | 4096 | 128 | 393827 |
197 | AVX Vectorized | 4096 | 128 | 185386 |
198 | AVX Vectorized FMA | 4096 | 128 | 187151 |
199 | Symmetry+Blocking | 4096 | 128 | 183065 |
200 | Blocking Optimized | 4096 | 128 | 124062 |
201 | Blocking Optimized & Adaptive | 4096 | 128 | 117631 |
202 | Baseline | 4096 | 256 | 1624158 |