search

hicder / muopdb

MuopDB - A Vector Database
https://github.com/hicder/muopdb/wiki
41 stars 3 forks source link

Introduce StreamingWithSIMDOptimized mode for PQ distance calculator #115

Closed hicder closed 5 days ago

hicder commented 5 days ago

With StreamingWithSIMDOptimized, in PQ, we will inline the SIMD calculator. This results in 2x performance improvement in most cases for PQ.

Benchmark:

PQ Distance/pq_distance_128_4_4/Scalar
                        time:   [111.75 ns 111.78 ns 111.82 ns]
Found 3 outliers among 100 measurements (3.00%)
  3 (3.00%) high mild
PQ Distance/pq_distance_128_4_4/SIMD
                        time:   [213.27 ns 213.38 ns 213.50 ns]
Found 5 outliers among 100 measurements (5.00%)
  5 (5.00%) high mild
PQ Distance/pq_distance_128_4_4/StreamingWithSIMD
                        time:   [95.488 ns 95.529 ns 95.572 ns]
Found 4 outliers among 100 measurements (4.00%)
  1 (1.00%) low mild
  2 (2.00%) high mild
  1 (1.00%) high severe
PQ Distance/pq_distance_128_4_4/StreamingWithSIMDOptimized
                        time:   [47.953 ns 48.041 ns 48.161 ns]
Found 10 outliers among 100 measurements (10.00%)
  2 (2.00%) high mild
  8 (8.00%) high severe
PQ Distance/pq_distance_128_4_8/Scalar
                        time:   [111.75 ns 111.79 ns 111.83 ns]
Found 6 outliers among 100 measurements (6.00%)
  2 (2.00%) low severe
  2 (2.00%) high mild
  2 (2.00%) high severe
PQ Distance/pq_distance_128_4_8/SIMD
                        time:   [213.01 ns 213.11 ns 213.22 ns]
Found 4 outliers among 100 measurements (4.00%)
  3 (3.00%) high mild
  1 (1.00%) high severe
PQ Distance/pq_distance_128_4_8/StreamingWithSIMD
                        time:   [95.836 ns 95.891 ns 95.946 ns]
Found 3 outliers among 100 measurements (3.00%)
  3 (3.00%) high mild
PQ Distance/pq_distance_128_4_8/StreamingWithSIMDOptimized
                        time:   [48.087 ns 48.115 ns 48.143 ns]
Found 8 outliers among 100 measurements (8.00%)
  1 (1.00%) low mild
  5 (5.00%) high mild
  2 (2.00%) high severe
PQ Distance/pq_distance_128_4_16/Scalar
                        time:   [112.38 ns 112.75 ns 113.04 ns]
PQ Distance/pq_distance_128_4_16/SIMD
                        time:   [215.37 ns 215.49 ns 215.62 ns]
Found 10 outliers among 100 measurements (10.00%)
  8 (8.00%) high mild
  2 (2.00%) high severe
PQ Distance/pq_distance_128_4_16/StreamingWithSIMD
                        time:   [96.036 ns 96.077 ns 96.123 ns]
Found 4 outliers among 100 measurements (4.00%)
  2 (2.00%) low mild
  1 (1.00%) high mild
  1 (1.00%) high severe
PQ Distance/pq_distance_128_4_16/StreamingWithSIMDOptimized
                        time:   [48.538 ns 48.605 ns 48.686 ns]
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high mild
PQ Distance/pq_distance_128_8_4/Scalar
                        time:   [66.942 ns 67.039 ns 67.125 ns]
Found 7 outliers among 100 measurements (7.00%)
  2 (2.00%) low severe
  3 (3.00%) low mild
  2 (2.00%) high mild
PQ Distance/pq_distance_128_8_4/SIMD
                        time:   [214.79 ns 214.85 ns 214.93 ns]
Found 14 outliers among 100 measurements (14.00%)
  1 (1.00%) low mild
  2 (2.00%) high mild
  11 (11.00%) high severe
PQ Distance/pq_distance_128_8_4/StreamingWithSIMD
                        time:   [56.940 ns 56.962 ns 56.985 ns]
Found 11 outliers among 100 measurements (11.00%)
  2 (2.00%) low severe
  5 (5.00%) low mild
  3 (3.00%) high mild
  1 (1.00%) high severe
PQ Distance/pq_distance_128_8_4/StreamingWithSIMDOptimized
                        time:   [29.395 ns 29.441 ns 29.489 ns]
Found 4 outliers among 100 measurements (4.00%)
  4 (4.00%) high mild
PQ Distance/pq_distance_128_8_8/Scalar
                        time:   [67.882 ns 68.211 ns 68.590 ns]
PQ Distance/pq_distance_128_8_8/SIMD
                        time:   [214.80 ns 214.85 ns 214.91 ns]
Found 18 outliers among 100 measurements (18.00%)
  3 (3.00%) high mild
  15 (15.00%) high severe
PQ Distance/pq_distance_128_8_8/StreamingWithSIMD
                        time:   [56.836 ns 56.866 ns 56.897 ns]
Found 7 outliers among 100 measurements (7.00%)
  2 (2.00%) low severe
  3 (3.00%) low mild
  1 (1.00%) high mild
  1 (1.00%) high severe
PQ Distance/pq_distance_128_8_8/StreamingWithSIMDOptimized
                        time:   [29.268 ns 29.306 ns 29.347 ns]
Found 4 outliers among 100 measurements (4.00%)
  4 (4.00%) high mild
PQ Distance/pq_distance_128_8_16/Scalar
                        time:   [66.077 ns 66.131 ns 66.188 ns]
Found 4 outliers among 100 measurements (4.00%)
  2 (2.00%) low severe
  1 (1.00%) low mild
  1 (1.00%) high mild
PQ Distance/pq_distance_128_8_16/SIMD
                        time:   [206.74 ns 206.88 ns 207.07 ns]
Found 9 outliers among 100 measurements (9.00%)
  3 (3.00%) high mild
  6 (6.00%) high severe
PQ Distance/pq_distance_128_8_16/StreamingWithSIMD
                        time:   [55.219 ns 55.304 ns 55.391 ns]
Found 9 outliers among 100 measurements (9.00%)
  7 (7.00%) low mild
  1 (1.00%) high mild
  1 (1.00%) high severe
PQ Distance/pq_distance_128_8_16/StreamingWithSIMDOptimized
                        time:   [28.727 ns 28.776 ns 28.826 ns]
Found 4 outliers among 100 measurements (4.00%)
  3 (3.00%) high mild
  1 (1.00%) high severe
PQ Distance/pq_distance_128_16_4/Scalar
                        time:   [53.886 ns 53.967 ns 54.052 ns]
Found 2 outliers among 100 measurements (2.00%)
  2 (2.00%) high mild
PQ Distance/pq_distance_128_16_4/SIMD
                        time:   [133.67 ns 133.71 ns 133.74 ns]
Found 7 outliers among 100 measurements (7.00%)
  1 (1.00%) low mild
  3 (3.00%) high mild
  3 (3.00%) high severe
PQ Distance/pq_distance_128_16_4/StreamingWithSIMD
                        time:   [41.210 ns 41.344 ns 41.488 ns]
Found 2 outliers among 100 measurements (2.00%)
  2 (2.00%) high mild
PQ Distance/pq_distance_128_16_4/StreamingWithSIMDOptimized
                        time:   [27.865 ns 27.949 ns 28.045 ns]
Found 15 outliers among 100 measurements (15.00%)
  8 (8.00%) high mild
  7 (7.00%) high severe
PQ Distance/pq_distance_128_16_8/Scalar
                        time:   [54.232 ns 54.313 ns 54.407 ns]
Found 4 outliers among 100 measurements (4.00%)
  4 (4.00%) high mild
PQ Distance/pq_distance_128_16_8/SIMD
                        time:   [133.60 ns 133.64 ns 133.69 ns]
Found 13 outliers among 100 measurements (13.00%)
  1 (1.00%) low severe
  2 (2.00%) low mild
  2 (2.00%) high mild
  8 (8.00%) high severe
PQ Distance/pq_distance_128_16_8/StreamingWithSIMD
                        time:   [41.559 ns 41.679 ns 41.799 ns]
Found 3 outliers among 100 measurements (3.00%)
  3 (3.00%) high mild
PQ Distance/pq_distance_128_16_8/StreamingWithSIMDOptimized
                        time:   [28.074 ns 28.225 ns 28.385 ns]
Found 6 outliers among 100 measurements (6.00%)
  6 (6.00%) high mild
PQ Distance/pq_distance_128_16_16/Scalar
                        time:   [53.951 ns 54.024 ns 54.095 ns]
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high mild
PQ Distance/pq_distance_128_16_16/SIMD
                        time:   [129.19 ns 129.23 ns 129.27 ns]
Found 8 outliers among 100 measurements (8.00%)
  3 (3.00%) high mild
  5 (5.00%) high severe
PQ Distance/pq_distance_128_16_16/StreamingWithSIMD
                        time:   [42.261 ns 42.312 ns 42.367 ns]
Found 3 outliers among 100 measurements (3.00%)
  2 (2.00%) high mild
  1 (1.00%) high severe
PQ Distance/pq_distance_128_16_16/StreamingWithSIMDOptimized
                        time:   [28.453 ns 28.542 ns 28.630 ns]
Found 4 outliers among 100 measurements (4.00%)
  4 (4.00%) low mild
PQ Distance/pq_distance_128_32_4/Scalar
                        time:   [63.255 ns 63.305 ns 63.362 ns]
Found 4 outliers among 100 measurements (4.00%)
  3 (3.00%) high mild
  1 (1.00%) high severe
PQ Distance/pq_distance_128_32_4/SIMD
                        time:   [63.086 ns 63.144 ns 63.209 ns]
Found 8 outliers among 100 measurements (8.00%)
  6 (6.00%) high mild
  2 (2.00%) high severe
PQ Distance/pq_distance_128_32_4/StreamingWithSIMD
                        time:   [27.536 ns 27.611 ns 27.689 ns]
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high mild
PQ Distance/pq_distance_128_32_4/StreamingWithSIMDOptimized
                        time:   [17.026 ns 17.156 ns 17.290 ns]
PQ Distance/pq_distance_128_32_8/Scalar
                        time:   [67.020 ns 67.048 ns 67.079 ns]
Found 5 outliers among 100 measurements (5.00%)
  3 (3.00%) high mild
  2 (2.00%) high severe
PQ Distance/pq_distance_128_32_8/SIMD
                        time:   [66.837 ns 66.869 ns 66.900 ns]
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high mild
PQ Distance/pq_distance_128_32_8/StreamingWithSIMD
                        time:   [34.449 ns 34.499 ns 34.549 ns]
Found 5 outliers among 100 measurements (5.00%)
  5 (5.00%) high mild
PQ Distance/pq_distance_128_32_8/StreamingWithSIMDOptimized
                        time:   [25.263 ns 25.303 ns 25.346 ns]
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high mild
PQ Distance/pq_distance_128_32_16/Scalar
                        time:   [66.733 ns 66.761 ns 66.791 ns]
Found 6 outliers among 100 measurements (6.00%)
  4 (4.00%) high mild
  2 (2.00%) high severe
PQ Distance/pq_distance_128_32_16/SIMD
                        time:   [66.378 ns 66.417 ns 66.456 ns]
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high mild
PQ Distance/pq_distance_128_32_16/StreamingWithSIMD
                        time:   [29.528 ns 29.622 ns 29.716 ns]
Found 2 outliers among 100 measurements (2.00%)
  1 (1.00%) low mild
  1 (1.00%) high mild
PQ Distance/pq_distance_128_32_16/StreamingWithSIMDOptimized
                        time:   [19.432 ns 19.665 ns 19.897 ns]
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high mild
PQ Distance/pq_distance_128_64_4/Scalar
                        time:   [32.272 ns 32.288 ns 32.305 ns]
Found 6 outliers among 100 measurements (6.00%)
  1 (1.00%) low mild
  5 (5.00%) high mild
PQ Distance/pq_distance_128_64_4/SIMD
                        time:   [34.098 ns 34.121 ns 34.146 ns]
Found 4 outliers among 100 measurements (4.00%)
  3 (3.00%) high mild
  1 (1.00%) high severe
PQ Distance/pq_distance_128_64_4/StreamingWithSIMD
                        time:   [20.305 ns 20.423 ns 20.541 ns]
Found 2 outliers among 100 measurements (2.00%)
  2 (2.00%) high mild
PQ Distance/pq_distance_128_64_4/StreamingWithSIMDOptimized
                        time:   [14.519 ns 14.953 ns 15.449 ns]
PQ Distance/pq_distance_128_64_8/Scalar
                        time:   [32.211 ns 32.227 ns 32.243 ns]
Found 6 outliers among 100 measurements (6.00%)
  1 (1.00%) low mild
  4 (4.00%) high mild
  1 (1.00%) high severe
PQ Distance/pq_distance_128_64_8/SIMD
                        time:   [34.045 ns 34.062 ns 34.079 ns]
Found 3 outliers among 100 measurements (3.00%)
  1 (1.00%) high mild
  2 (2.00%) high severe
PQ Distance/pq_distance_128_64_8/StreamingWithSIMD
                        time:   [20.404 ns 20.503 ns 20.607 ns]
Found 5 outliers among 100 measurements (5.00%)
  3 (3.00%) low mild
  2 (2.00%) high mild
PQ Distance/pq_distance_128_64_8/StreamingWithSIMDOptimized
                        time:   [15.654 ns 16.234 ns 16.879 ns]
PQ Distance/pq_distance_128_64_16/Scalar
                        time:   [31.293 ns 31.316 ns 31.338 ns]
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high mild
PQ Distance/pq_distance_128_64_16/SIMD
                        time:   [33.303 ns 33.350 ns 33.399 ns]
Found 5 outliers among 100 measurements (5.00%)
  5 (5.00%) high mild
PQ Distance/pq_distance_128_64_16/StreamingWithSIMD
                        time:   [22.811 ns 22.922 ns 23.044 ns]
Found 4 outliers among 100 measurements (4.00%)
  1 (1.00%) high mild
  3 (3.00%) high severe
PQ Distance/pq_distance_128_64_16/StreamingWithSIMDOptimized
                        time:   [17.949 ns 18.738 ns 19.576 ns]
PQ Distance/pq_distance_128_128_4/Scalar
                        time:   [19.348 ns 19.433 ns 19.527 ns]
Found 4 outliers among 100 measurements (4.00%)
  4 (4.00%) high mild
PQ Distance/pq_distance_128_128_4/SIMD
                        time:   [22.293 ns 22.362 ns 22.438 ns]
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high mild
PQ Distance/pq_distance_128_128_4/StreamingWithSIMD
                        time:   [20.037 ns 20.098 ns 20.161 ns]
Found 3 outliers among 100 measurements (3.00%)
  3 (3.00%) high mild
PQ Distance/pq_distance_128_128_4/StreamingWithSIMDOptimized
                        time:   [11.898 ns 11.948 ns 12.005 ns]
Found 8 outliers among 100 measurements (8.00%)
  6 (6.00%) high mild
  2 (2.00%) high severe
PQ Distance/pq_distance_128_128_8/Scalar
                        time:   [31.731 ns 31.762 ns 31.794 ns]
Found 3 outliers among 100 measurements (3.00%)
  3 (3.00%) high mild
PQ Distance/pq_distance_128_128_8/SIMD
                        time:   [31.631 ns 31.684 ns 31.743 ns]
Found 5 outliers among 100 measurements (5.00%)
  4 (4.00%) high mild
  1 (1.00%) high severe
PQ Distance/pq_distance_128_128_8/StreamingWithSIMD
                        time:   [27.717 ns 27.750 ns 27.786 ns]
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high mild
PQ Distance/pq_distance_128_128_8/StreamingWithSIMDOptimized
                        time:   [23.095 ns 23.120 ns 23.147 ns]
Found 2 outliers among 100 measurements (2.00%)
  2 (2.00%) high mild
PQ Distance/pq_distance_128_128_16/Scalar
                        time:   [20.898 ns 20.921 ns 20.946 ns]
Found 7 outliers among 100 measurements (7.00%)
  2 (2.00%) low mild
  5 (5.00%) high mild
PQ Distance/pq_distance_128_128_16/SIMD
                        time:   [23.345 ns 23.508 ns 23.661 ns]
PQ Distance/pq_distance_128_128_16/StreamingWithSIMD
                        time:   [22.527 ns 22.568 ns 22.607 ns]
Found 7 outliers among 100 measurements (7.00%)
  2 (2.00%) low severe
  4 (4.00%) low mild
  1 (1.00%) high severe
PQ Distance/pq_distance_128_128_16/StreamingWithSIMDOptimized
                        time:   [13.975 ns 14.117 ns 14.270 ns]
Found 4 outliers among 100 measurements (4.00%)
  4 (4.00%) high mild
PQ Distance/pq_distance_256_4_4/Scalar
                        time:   [209.05 ns 209.10 ns 209.16 ns]
Found 4 outliers among 100 measurements (4.00%)
  1 (1.00%) high mild
  3 (3.00%) high severe
PQ Distance/pq_distance_256_4_4/SIMD
                        time:   [408.44 ns 408.71 ns 408.99 ns]
Found 3 outliers among 100 measurements (3.00%)
  2 (2.00%) high mild
  1 (1.00%) high severe
PQ Distance/pq_distance_256_4_4/StreamingWithSIMD
                        time:   [175.01 ns 175.07 ns 175.13 ns]
Found 3 outliers among 100 measurements (3.00%)
  1 (1.00%) high mild
  2 (2.00%) high severe
PQ Distance/pq_distance_256_4_4/StreamingWithSIMDOptimized
                        time:   [86.214 ns 86.516 ns 86.879 ns]
PQ Distance/pq_distance_256_4_8/Scalar
                        time:   [215.74 ns 215.78 ns 215.83 ns]
Found 11 outliers among 100 measurements (11.00%)
  2 (2.00%) low severe
  3 (3.00%) high mild
  6 (6.00%) high severe
PQ Distance/pq_distance_256_4_8/SIMD
                        time:   [408.05 ns 408.23 ns 408.43 ns]
Found 14 outliers among 100 measurements (14.00%)
  6 (6.00%) high mild
  8 (8.00%) high severe
PQ Distance/pq_distance_256_4_8/StreamingWithSIMD
                        time:   [174.98 ns 175.02 ns 175.07 ns]
Found 11 outliers among 100 measurements (11.00%)
  3 (3.00%) low mild
  3 (3.00%) high mild
  5 (5.00%) high severe
PQ Distance/pq_distance_256_4_8/StreamingWithSIMDOptimized
                        time:   [89.142 ns 89.434 ns 89.686 ns]
Found 11 outliers among 100 measurements (11.00%)
  6 (6.00%) low severe
  3 (3.00%) low mild
  2 (2.00%) high mild
PQ Distance/pq_distance_256_4_16/Scalar
                        time:   [218.03 ns 218.12 ns 218.23 ns]
Found 3 outliers among 100 measurements (3.00%)
  1 (1.00%) high mild
  2 (2.00%) high severe
PQ Distance/pq_distance_256_4_16/SIMD
                        time:   [414.44 ns 414.73 ns 415.05 ns]
Found 5 outliers among 100 measurements (5.00%)
  4 (4.00%) high mild
  1 (1.00%) high severe
PQ Distance/pq_distance_256_4_16/StreamingWithSIMD
                        time:   [176.47 ns 176.57 ns 176.68 ns]
Found 3 outliers among 100 measurements (3.00%)
  1 (1.00%) low mild
  2 (2.00%) high severe
PQ Distance/pq_distance_256_4_16/StreamingWithSIMDOptimized
                        time:   [89.714 ns 89.792 ns 89.860 ns]
Found 8 outliers among 100 measurements (8.00%)
  2 (2.00%) low severe
  2 (2.00%) low mild
  2 (2.00%) high mild
  2 (2.00%) high severe
PQ Distance/pq_distance_256_8_4/Scalar
                        time:   [123.45 ns 123.50 ns 123.56 ns]
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high mild
PQ Distance/pq_distance_256_8_4/SIMD
                        time:   [403.73 ns 403.85 ns 403.99 ns]
Found 8 outliers among 100 measurements (8.00%)
  5 (5.00%) high mild
  3 (3.00%) high severe
PQ Distance/pq_distance_256_8_4/StreamingWithSIMD
                        time:   [94.539 ns 94.603 ns 94.661 ns]
Found 8 outliers among 100 measurements (8.00%)
  6 (6.00%) low mild
  2 (2.00%) high mild
PQ Distance/pq_distance_256_8_4/StreamingWithSIMDOptimized
                        time:   [49.279 ns 49.342 ns 49.408 ns]
Found 7 outliers among 100 measurements (7.00%)
  5 (5.00%) low mild
  2 (2.00%) high mild
PQ Distance/pq_distance_256_8_8/Scalar
                        time:   [123.58 ns 123.64 ns 123.70 ns]
PQ Distance/pq_distance_256_8_8/SIMD
                        time:   [400.15 ns 400.26 ns 400.37 ns]
Found 9 outliers among 100 measurements (9.00%)
  2 (2.00%) high mild
  7 (7.00%) high severe
PQ Distance/pq_distance_256_8_8/StreamingWithSIMD
                        time:   [94.156 ns 94.218 ns 94.285 ns]
Found 4 outliers among 100 measurements (4.00%)
  1 (1.00%) low mild
  2 (2.00%) high mild
  1 (1.00%) high severe
PQ Distance/pq_distance_256_8_8/StreamingWithSIMDOptimized
                        time:   [48.542 ns 48.596 ns 48.652 ns]
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high mild
PQ Distance/pq_distance_256_8_16/Scalar
                        time:   [124.66 ns 124.72 ns 124.78 ns]
PQ Distance/pq_distance_256_8_16/SIMD
                        time:   [406.10 ns 406.32 ns 406.60 ns]
Found 2 outliers among 100 measurements (2.00%)
  2 (2.00%) high mild
PQ Distance/pq_distance_256_8_16/StreamingWithSIMD
                        time:   [98.313 ns 98.348 ns 98.387 ns]
Found 11 outliers among 100 measurements (11.00%)
  2 (2.00%) low severe
  1 (1.00%) low mild
  7 (7.00%) high mild
  1 (1.00%) high severe
PQ Distance/pq_distance_256_8_16/StreamingWithSIMDOptimized
                        time:   [50.778 ns 50.824 ns 50.873 ns]
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) low mild
PQ Distance/pq_distance_256_16_4/Scalar
                        time:   [104.12 ns 104.23 ns 104.34 ns]
Found 2 outliers among 100 measurements (2.00%)
  2 (2.00%) high mild
PQ Distance/pq_distance_256_16_4/SIMD
                        time:   [262.32 ns 264.12 ns 265.78 ns]
Found 13 outliers among 100 measurements (13.00%)
  12 (12.00%) high mild
  1 (1.00%) high severe
PQ Distance/pq_distance_256_16_4/StreamingWithSIMD
                        time:   [65.708 ns 65.847 ns 66.001 ns]
Found 2 outliers among 100 measurements (2.00%)
  2 (2.00%) high mild
PQ Distance/pq_distance_256_16_4/StreamingWithSIMDOptimized
                        time:   [42.713 ns 42.809 ns 42.907 ns]
Found 5 outliers among 100 measurements (5.00%)
  5 (5.00%) high mild
PQ Distance/pq_distance_256_16_8/Scalar
                        time:   [102.64 ns 102.73 ns 102.82 ns]
Found 5 outliers among 100 measurements (5.00%)
  5 (5.00%) high mild
PQ Distance/pq_distance_256_16_8/SIMD
                        time:   [258.57 ns 258.65 ns 258.74 ns]
Found 12 outliers among 100 measurements (12.00%)
  1 (1.00%) low mild
  4 (4.00%) high mild
  7 (7.00%) high severe
PQ Distance/pq_distance_256_16_8/StreamingWithSIMD
                        time:   [62.677 ns 62.779 ns 62.883 ns]
PQ Distance/pq_distance_256_16_8/StreamingWithSIMDOptimized
                        time:   [41.550 ns 41.671 ns 41.785 ns]
PQ Distance/pq_distance_256_16_16/Scalar
                        time:   [100.43 ns 100.53 ns 100.64 ns]
Found 3 outliers among 100 measurements (3.00%)
  3 (3.00%) high mild
PQ Distance/pq_distance_256_16_16/SIMD
                        time:   [261.29 ns 261.40 ns 261.55 ns]
Found 8 outliers among 100 measurements (8.00%)
  2 (2.00%) high mild
  6 (6.00%) high severe
PQ Distance/pq_distance_256_16_16/StreamingWithSIMD
                        time:   [65.905 ns 66.046 ns 66.186 ns]
Found 6 outliers among 100 measurements (6.00%)
  2 (2.00%) low mild
  4 (4.00%) high mild
PQ Distance/pq_distance_256_16_16/StreamingWithSIMDOptimized
                        time:   [42.725 ns 42.851 ns 42.976 ns]
Found 3 outliers among 100 measurements (3.00%)
  2 (2.00%) low mild
  1 (1.00%) high mild
PQ Distance/pq_distance_256_32_4/Scalar
                        time:   [133.14 ns 133.20 ns 133.27 ns]
Found 8 outliers among 100 measurements (8.00%)
  3 (3.00%) high mild
  5 (5.00%) high severe
PQ Distance/pq_distance_256_32_4/SIMD
                        time:   [132.55 ns 132.72 ns 132.91 ns]
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high mild
PQ Distance/pq_distance_256_32_4/StreamingWithSIMD
                        time:   [47.306 ns 47.444 ns 47.564 ns]
Found 4 outliers among 100 measurements (4.00%)
  1 (1.00%) low severe
  2 (2.00%) low mild
  1 (1.00%) high mild
PQ Distance/pq_distance_256_32_4/StreamingWithSIMDOptimized
                        time:   [33.471 ns 33.635 ns 33.800 ns]
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high mild
PQ Distance/pq_distance_256_32_8/Scalar
                        time:   [132.66 ns 132.72 ns 132.78 ns]
Found 8 outliers among 100 measurements (8.00%)
  6 (6.00%) high mild
  2 (2.00%) high severe
PQ Distance/pq_distance_256_32_8/SIMD
                        time:   [131.43 ns 131.49 ns 131.56 ns]
Found 9 outliers among 100 measurements (9.00%)
  4 (4.00%) high mild
  5 (5.00%) high severe
PQ Distance/pq_distance_256_32_8/StreamingWithSIMD
                        time:   [47.014 ns 47.125 ns 47.239 ns]
PQ Distance/pq_distance_256_32_8/StreamingWithSIMDOptimized
                        time:   [33.018 ns 33.289 ns 33.566 ns]
PQ Distance/pq_distance_256_32_16/Scalar
                        time:   [134.78 ns 134.83 ns 134.88 ns]
Found 8 outliers among 100 measurements (8.00%)
  6 (6.00%) high mild
  2 (2.00%) high severe
PQ Distance/pq_distance_256_32_16/SIMD
                        time:   [132.72 ns 132.78 ns 132.87 ns]
Found 5 outliers among 100 measurements (5.00%)
  2 (2.00%) high mild
  3 (3.00%) high severe
PQ Distance/pq_distance_256_32_16/StreamingWithSIMD
                        time:   [48.935 ns 49.024 ns 49.111 ns]
PQ Distance/pq_distance_256_32_16/StreamingWithSIMDOptimized
                        time:   [33.833 ns 34.123 ns 34.411 ns]
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) low mild
PQ Distance/pq_distance_256_64_4/Scalar
                        time:   [70.670 ns 70.721 ns 70.774 ns]
Found 2 outliers among 100 measurements (2.00%)
  2 (2.00%) high mild
PQ Distance/pq_distance_256_64_4/SIMD
                        time:   [70.441 ns 70.473 ns 70.508 ns]
Found 2 outliers among 100 measurements (2.00%)
  1 (1.00%) high mild
  1 (1.00%) high severe
PQ Distance/pq_distance_256_64_4/StreamingWithSIMD
                        time:   [38.372 ns 38.424 ns 38.477 ns]
Found 5 outliers among 100 measurements (5.00%)
  4 (4.00%) low mild
  1 (1.00%) high mild
PQ Distance/pq_distance_256_64_4/StreamingWithSIMDOptimized
                        time:   [23.068 ns 23.190 ns 23.310 ns]
Found 3 outliers among 100 measurements (3.00%)
  2 (2.00%) low mild
  1 (1.00%) high mild
PQ Distance/pq_distance_256_64_8/Scalar
                        time:   [70.654 ns 70.701 ns 70.750 ns]
Found 3 outliers among 100 measurements (3.00%)
  2 (2.00%) high mild
  1 (1.00%) high severe
PQ Distance/pq_distance_256_64_8/SIMD
                        time:   [70.579 ns 70.620 ns 70.662 ns]
Found 8 outliers among 100 measurements (8.00%)
  2 (2.00%) low mild
  5 (5.00%) high mild
  1 (1.00%) high severe
PQ Distance/pq_distance_256_64_8/StreamingWithSIMD
                        time:   [37.805 ns 37.864 ns 37.923 ns]
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high mild
PQ Distance/pq_distance_256_64_8/StreamingWithSIMDOptimized
                        time:   [24.155 ns 24.253 ns 24.357 ns]
Found 2 outliers among 100 measurements (2.00%)
  1 (1.00%) high mild
  1 (1.00%) high severe
PQ Distance/pq_distance_256_64_16/Scalar
                        time:   [68.409 ns 68.464 ns 68.524 ns]
Found 2 outliers among 100 measurements (2.00%)
  2 (2.00%) high mild
PQ Distance/pq_distance_256_64_16/SIMD
                        time:   [68.118 ns 68.178 ns 68.241 ns]
Found 8 outliers among 100 measurements (8.00%)
  7 (7.00%) high mild
  1 (1.00%) high severe
PQ Distance/pq_distance_256_64_16/StreamingWithSIMD
                        time:   [38.220 ns 38.293 ns 38.366 ns]
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high severe
PQ Distance/pq_distance_256_64_16/StreamingWithSIMDOptimized
                        time:   [24.347 ns 24.462 ns 24.581 ns]
Found 3 outliers among 100 measurements (3.00%)
  1 (1.00%) low mild
  2 (2.00%) high mild
PQ Distance/pq_distance_256_128_4/Scalar
                        time:   [35.418 ns 35.469 ns 35.519 ns]
Found 5 outliers among 100 measurements (5.00%)
  5 (5.00%) high mild
PQ Distance/pq_distance_256_128_4/SIMD
                        time:   [35.168 ns 35.247 ns 35.331 ns]
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high mild
PQ Distance/pq_distance_256_128_4/StreamingWithSIMD
                        time:   [33.506 ns 33.956 ns 34.395 ns]
Found 6 outliers among 100 measurements (6.00%)
  6 (6.00%) low mild
PQ Distance/pq_distance_256_128_4/StreamingWithSIMDOptimized
                        time:   [23.185 ns 23.253 ns 23.318 ns]
Found 5 outliers among 100 measurements (5.00%)
  2 (2.00%) low severe
  3 (3.00%) low mild
PQ Distance/pq_distance_256_128_8/Scalar
                        time:   [36.406 ns 36.462 ns 36.524 ns]
Found 2 outliers among 100 measurements (2.00%)
  2 (2.00%) high mild
PQ Distance/pq_distance_256_128_8/SIMD
                        time:   [36.213 ns 36.265 ns 36.318 ns]
Found 4 outliers among 100 measurements (4.00%)
  2 (2.00%) high mild
  2 (2.00%) high severe
PQ Distance/pq_distance_256_128_8/StreamingWithSIMD
                        time:   [30.876 ns 32.099 ns 33.236 ns]
PQ Distance/pq_distance_256_128_8/StreamingWithSIMDOptimized
                        time:   [23.784 ns 23.929 ns 24.073 ns]
PQ Distance/pq_distance_256_128_16/Scalar
                        time:   [36.807 ns 36.846 ns 36.885 ns]
PQ Distance/pq_distance_256_128_16/SIMD
                        time:   [37.935 ns 37.972 ns 38.009 ns]
Found 5 outliers among 100 measurements (5.00%)
  2 (2.00%) low mild
  3 (3.00%) high mild
PQ Distance/pq_distance_256_128_16/StreamingWithSIMD
                        time:   [31.003 ns 31.569 ns 32.101 ns]
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high mild
PQ Distance/pq_distance_256_128_16/StreamingWithSIMDOptimized
                        time:   [25.266 ns 25.332 ns 25.401 ns]
Found 4 outliers among 100 measurements (4.00%)
  3 (3.00%) low mild
  1 (1.00%) high mild