feat(cli): optimize buffer size for highest throughput #16
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Profiling Reportencode/3 time: [28.882 ns 28.898 ns 28.917 ns]
thrpt: [98.938 MiB/s 99.005 MiB/s 99.060 MiB/s]
change:
time: [-1.0145% -0.8177% -0.5413%] (p = 0.00 < 0.05)
thrpt: [+0.5442% +0.8244% +1.0249%]
Change within noise threshold.
Found 11 outliers among 100 measurements (11.00%)
1 (1.00%) low severe
7 (7.00%) low mild
1 (1.00%) high mild
2 (2.00%) high severe
encode/50 time: [62.362 ns 62.477 ns 62.614 ns]
thrpt: [761.54 MiB/s 763.23 MiB/s 764.63 MiB/s]
change:
time: [-1.4424% -1.2133% -0.9697%] (p = 0.00 < 0.05)
thrpt: [+0.9792% +1.2282% +1.4636%]
Change within noise threshold.
Found 10 outliers among 100 measurements (10.00%)
4 (4.00%) low mild
3 (3.00%) high mild
3 (3.00%) high severe
encode/100 time: [108.20 ns 108.35 ns 108.52 ns]
thrpt: [878.82 MiB/s 880.14 MiB/s 881.43 MiB/s]
change:
time: [-0.9126% -0.7036% -0.4994%] (p = 0.00 < 0.05)
thrpt: [+0.5019% +0.7086% +0.9210%]
Change within noise threshold.
Found 1 outliers among 100 measurements (1.00%)
1 (1.00%) high severe
encode/500 time: [348.03 ns 348.37 ns 348.69 ns]
thrpt: [1.3355 GiB/s 1.3367 GiB/s 1.3380 GiB/s]
change:
time: [-0.3977% -0.2822% -0.1592%] (p = 0.00 < 0.05)
thrpt: [+0.1595% +0.2830% +0.3993%]
Change within noise threshold.
Found 1 outliers among 100 measurements (1.00%)
1 (1.00%) low mild
encode/3072 time: [1.9692 µs 1.9699 µs 1.9710 µs]
thrpt: [1.4516 GiB/s 1.4524 GiB/s 1.4529 GiB/s]
change:
time: [+0.1591% +0.2919% +0.4812%] (p = 0.00 < 0.05)
thrpt: [-0.4789% -0.2911% -0.1588%]
Change within noise threshold.
Found 10 outliers among 100 measurements (10.00%)
1 (1.00%) low mild
3 (3.00%) high mild
6 (6.00%) high severe
encode/51200 time: [32.188 µs 32.192 µs 32.197 µs]
thrpt: [1.4810 GiB/s 1.4812 GiB/s 1.4814 GiB/s]
change:
time: [-0.0661% -0.0411% -0.0179%] (p = 0.00 < 0.05)
thrpt: [+0.0179% +0.0412% +0.0661%]
Change within noise threshold.
Found 9 outliers among 100 measurements (9.00%)
2 (2.00%) low severe
3 (3.00%) high mild
4 (4.00%) high severe
encode/102400 time: [64.406 µs 64.417 µs 64.436 µs]
thrpt: [1.4800 GiB/s 1.4805 GiB/s 1.4807 GiB/s]
change:
time: [-0.0479% -0.0336% -0.0192%] (p = 0.00 < 0.05)
thrpt: [+0.0192% +0.0336% +0.0479%]
Change within noise threshold.
Found 9 outliers among 100 measurements (9.00%)
3 (3.00%) low mild
3 (3.00%) high mild
3 (3.00%) high severe
encode/512000 time: [208.71 µs 214.05 µs 220.32 µs]
thrpt: [2.1642 GiB/s 2.2277 GiB/s 2.2847 GiB/s]
change:
time: [-12.888% -5.8590% +1.4798%] (p = 0.12 > 0.05)
thrpt: [-1.4582% +6.2236% +14.794%]
No change in performance detected.
Found 25 outliers among 100 measurements (25.00%)
2 (2.00%) high mild
23 (23.00%) high severe
encode/1048576 time: [444.42 µs 455.81 µs 468.43 µs]
thrpt: [2.0848 GiB/s 2.1425 GiB/s 2.1974 GiB/s]
change:
time: [-3.3281% +2.4821% +8.3518%] (p = 0.42 > 0.05)
thrpt: [-7.7081% -2.4220% +3.4427%]
No change in performance detected.
Found 3 outliers among 100 measurements (3.00%)
3 (3.00%) high mild
encode/5242880 time: [2.2158 ms 2.2736 ms 2.3323 ms]
thrpt: [2.0935 GiB/s 2.1476 GiB/s 2.2036 GiB/s]
change:
time: [-2.9679% +0.4126% +3.7148%] (p = 0.81 > 0.05)
thrpt: [-3.5817% -0.4110% +3.0587%]
No change in performance detected.
Found 1 outliers among 100 measurements (1.00%)
1 (1.00%) high mild
encode/10485760 time: [3.9766 ms 4.0195 ms 4.0742 ms]
thrpt: [2.3970 GiB/s 2.4296 GiB/s 2.4558 GiB/s]
change:
time: [+0.1102% +1.2593% +3.0845%] (p = 0.06 > 0.05)
thrpt: [-2.9922% -1.2436% -0.1101%]
No change in performance detected.
Found 13 outliers among 100 measurements (13.00%)
1 (1.00%) low mild
5 (5.00%) high mild
7 (7.00%) high severe
encode/20971520 time: [8.1901 ms 8.2991 ms 8.4305 ms]
thrpt: [2.3167 GiB/s 2.3534 GiB/s 2.3847 GiB/s]
change:
time: [-3.5959% -1.2780% +1.1085%] (p = 0.30 > 0.05)
thrpt: [-1.0964% +1.2945% +3.7300%]
No change in performance detected.
Found 5 outliers among 100 measurements (5.00%)
1 (1.00%) high mild
4 (4.00%) high severe
decode/3 time: [21.743 ns 21.771 ns 21.801 ns]
thrpt: [131.23 MiB/s 131.41 MiB/s 131.59 MiB/s]
change:
+ time: [-3.0858% -2.6671% -2.2662%] (p = 0.00 < 0.05)
+ thrpt: [+2.3188% +2.7402% +3.1840%]
+ Performance has improved.
Found 3 outliers among 100 measurements (3.00%)
1 (1.00%) low mild
2 (2.00%) high mild
decode/50 time: [46.679 ns 46.878 ns 47.109 ns]
thrpt: [1012.2 MiB/s 1017.2 MiB/s 1021.5 MiB/s]
change:
+ time: [-5.7175% -5.2229% -4.7702%] (p = 0.00 < 0.05)
+ thrpt: [+5.0091% +5.5107% +6.0642%]
+ Performance has improved.
Found 21 outliers among 100 measurements (21.00%)
2 (2.00%) low severe
4 (4.00%) low mild
6 (6.00%) high mild
9 (9.00%) high severe
decode/100 time: [78.298 ns 78.541 ns 78.816 ns]
thrpt: [1.1816 GiB/s 1.1858 GiB/s 1.1895 GiB/s]
change:
+ time: [-4.2022% -3.8974% -3.5392%] (p = 0.00 < 0.05)
+ thrpt: [+3.6691% +4.0555% +4.3866%]
+ Performance has improved.
Found 15 outliers among 100 measurements (15.00%)
3 (3.00%) low severe
1 (1.00%) low mild
5 (5.00%) high mild
6 (6.00%) high severe
decode/500 time: [302.88 ns 303.94 ns 305.00 ns]
thrpt: [1.5267 GiB/s 1.5321 GiB/s 1.5374 GiB/s]
change:
time: [-0.4847% +2.6655% +6.9281%] (p = 0.15 > 0.05)
thrpt: [-6.4792% -2.5963% +0.4870%]
No change in performance detected.
Found 5 outliers among 100 measurements (5.00%)
1 (1.00%) high mild
4 (4.00%) high severe
decode/3072 time: [1.7015 µs 1.7018 µs 1.7022 µs]
thrpt: [1.6808 GiB/s 1.6811 GiB/s 1.6814 GiB/s]
change:
+ time: [-1.4296% -1.3231% -1.2472%] (p = 0.00 < 0.05)
+ thrpt: [+1.2629% +1.3409% +1.4503%]
+ Performance has improved.
Found 2 outliers among 100 measurements (2.00%)
1 (1.00%) low mild
1 (1.00%) high severe
decode/51200 time: [27.618 µs 27.620 µs 27.622 µs]
thrpt: [1.7263 GiB/s 1.7264 GiB/s 1.7265 GiB/s]
change:
time: [-0.1383% -0.1162% -0.0956%] (p = 0.00 < 0.05)
thrpt: [+0.0957% +0.1164% +0.1385%]
Change within noise threshold.
Found 8 outliers among 100 measurements (8.00%)
4 (4.00%) low mild
2 (2.00%) high mild
2 (2.00%) high severe
decode/102400 time: [55.533 µs 55.539 µs 55.546 µs]
thrpt: [1.7169 GiB/s 1.7171 GiB/s 1.7173 GiB/s]
change:
time: [-0.1454% -0.1270% -0.1094%] (p = 0.00 < 0.05)
thrpt: [+0.1095% +0.1271% +0.1456%]
Change within noise threshold.
Found 8 outliers among 100 measurements (8.00%)
2 (2.00%) low severe
1 (1.00%) low mild
3 (3.00%) high mild
2 (2.00%) high severe
decode/512000 time: [162.36 µs 165.34 µs 169.08 µs]
thrpt: [2.8202 GiB/s 2.8840 GiB/s 2.9369 GiB/s]
change:
time: [-7.5955% -1.1314% +4.8265%] (p = 0.75 > 0.05)
thrpt: [-4.6043% +1.1443% +8.2198%]
No change in performance detected.
Found 19 outliers among 100 measurements (19.00%)
1 (1.00%) high mild
18 (18.00%) high severe
decode/1048576 time: [344.79 µs 354.06 µs 364.36 µs]
thrpt: [2.6802 GiB/s 2.7582 GiB/s 2.8323 GiB/s]
change:
+ time: [-13.166% -7.1943% -1.1726%] (p = 0.03 < 0.05)
+ thrpt: [+1.1865% +7.7521% +15.162%]
+ Performance has improved.
Found 10 outliers among 100 measurements (10.00%)
10 (10.00%) high mild
decode/5242880 time: [1.8217 ms 1.8718 ms 1.9223 ms]
thrpt: [2.5401 GiB/s 2.6086 GiB/s 2.6803 GiB/s]
change:
time: [-1.9101% +1.6180% +5.5347%] (p = 0.40 > 0.05)
thrpt: [-5.2444% -1.5923% +1.9473%]
No change in performance detected.
decode/10485760 time: [3.1083 ms 3.1348 ms 3.1656 ms]
thrpt: [3.0850 GiB/s 3.1152 GiB/s 3.1417 GiB/s]
change:
time: [-1.6372% -0.4491% +0.7634%] (p = 0.48 > 0.05)
thrpt: [-0.7576% +0.4511% +1.6644%]
No change in performance detected.
Found 8 outliers among 100 measurements (8.00%)
7 (7.00%) high mild
1 (1.00%) high severe
decode/20971520 time: [6.2172 ms 6.3003 ms 6.4058 ms]
thrpt: [3.0490 GiB/s 3.1001 GiB/s 3.1415 GiB/s]
change:
time: [-3.1880% -0.7652% +1.7529%] (p = 0.54 > 0.05)
thrpt: [-1.7227% +0.7711% +3.2930%]
No change in performance detected.
Found 11 outliers among 100 measurements (11.00%)
2 (2.00%) high mild
9 (9.00%) high severe
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
From tests, ~5 KB buffer size provides about optimal CLI throughput. Additional buffer-size throughput gains require much larger buffer sizes for little improvement.