Re-implement io.pedestal.interceptor.chain #841
Draft
+955
−460
Conversation
This file contains 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
v1 is beating v2 most of the time, that's not good. ┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━━┳━━━━━━━━━━━┓ ┃ Expression ┃ Mean ┃ Var ┃ Ratio ┃ ┣━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╋━━━━━━━━━━━╋━━━━━━━━━━━━━╋━━━━━━━━━━━┫ ┃ (example chain1/execute true 10) ┃ 294.73 µs ┃ ± 17.88 µs ┃ 126.2 % ┃ ┃ (example chain2/execute true 10) ┃ 233.47 µs ┃ ± 6.49 µs ┃ 100.0 % ┃ (fastest) ┃ (example chain1/execute true 50) ┃ 278.71 µs ┃ ± 7.70 µs ┃ 119.4 % ┃ ┃ (example chain2/execute true 50) ┃ 2.35 ms ┃ ± 127.43 µs ┃ 1,005.9 % ┃ (slowest) ┃ (example chain1/execute true 100) ┃ 237.70 µs ┃ ± 7.42 µs ┃ 101.8 % ┃ ┃ (example chain2/execute true 100) ┃ 1.36 ms ┃ ± 620.82 µs ┃ 581.7 % ┃ ┃ (example chain1/execute false 10) ┃ 1.20 ms ┃ ± 110.45 µs ┃ 514.6 % ┃ ┃ (example chain2/execute false 10) ┃ 1.86 ms ┃ ± 59.25 µs ┃ 798.8 % ┃ ┃ (example chain1/execute false 50) ┃ 953.14 µs ┃ ± 26.90 µs ┃ 408.3 % ┃ ┃ (example chain2/execute false 50) ┃ 2.16 ms ┃ ± 134.51 µs ┃ 927.0 % ┃ ┃ (example chain1/execute false 100) ┃ 1.89 ms ┃ ± 62.11 µs ┃ 810.6 % ┃ ┃ (example chain2/execute false 100) ┃ 991.04 µs ┃ ± 53.19 µs ┃ 424.5 % ┃ ┗━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┻━━━━━━━━━━━┻━━━━━━━━━━━━━┻━━━━━━━━━━━┛
feels like PersistentQueue may be the way to go after all, gettting nth element of a vector is too slow. ┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━━┳━━━━━━━━━┓ ┃ Expression ┃ Mean ┃ Var ┃ Ratio ┃ ┣━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╋━━━━━━━━━━━╋━━━━━━━━━━━━━╋━━━━━━━━━┫ ┃ (example chain1/execute true 10) ┃ 299.00 µs ┃ ± 11.98 µs ┃ 128.7 % ┃ ┃ (example chain2/execute true 10) ┃ 235.72 µs ┃ ± 6.02 µs ┃ 101.5 % ┃ ┃ (example chain1/execute true 50) ┃ 271.39 µs ┃ ± 2.73 µs ┃ 116.9 % ┃ ┃ (example chain2/execute true 50) ┃ 2.23 ms ┃ ± 60.40 µs ┃ 959.1 % ┃ ┃ (example chain1/execute true 100) ┃ 232.24 µs ┃ ± 653.58 ns ┃ 100.0 % ┃ (fastest) ┃ (example chain2/execute true 100) ┃ 1.13 ms ┃ ± 40.31 µs ┃ 487.2 % ┃ ┃ (example chain1/execute false 10) ┃ 1.19 ms ┃ ± 43.36 µs ┃ 510.9 % ┃ ┃ (example chain2/execute false 10) ┃ 1.85 ms ┃ ± 50.32 µs ┃ 797.6 % ┃ ┃ (example chain1/execute false 50) ┃ 964.75 µs ┃ ± 29.27 µs ┃ 415.4 % ┃ ┃ (example chain2/execute false 50) ┃ 2.23 ms ┃ ± 46.00 µs ┃ 961.8 % ┃ (slowest) ┃ (example chain1/execute false 100) ┃ 1.89 ms ┃ ± 52.39 µs ┃ 815.1 % ┃ ┃ (example chain2/execute false 100) ┃ 973.70 µs ┃ ± 28.41 µs ┃ 419.3 % ┃ ┗━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┻━━━━━━━━━━━┻━━━━━━━━━━━━━┻━━━━━━━━━┛
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━━┳━━━━━━━━━┓ ┃ Expression ┃ Mean ┃ Var ┃ Ratio ┃ ┣━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╋━━━━━━━━━━━╋━━━━━━━━━━━━━╋━━━━━━━━━┫ ┃ (example chain1/execute true 10) ┃ 300.21 µs ┃ ± 12.59 µs ┃ 114.0 % ┃ ┃ (example chain2/execute true 10) ┃ 263.24 µs ┃ ± 11.35 µs ┃ 100.0 % ┃ (fastest) ┃ (example chain1/execute true 50) ┃ 290.43 µs ┃ ± 15.45 µs ┃ 110.3 % ┃ ┃ (example chain2/execute true 50) ┃ 2.16 ms ┃ ± 142.63 µs ┃ 821.2 % ┃ (slowest) ┃ (example chain1/execute true 100) ┃ 265.68 µs ┃ ± 9.36 µs ┃ 100.9 % ┃ ┃ (example chain2/execute true 100) ┃ 1.08 ms ┃ ± 41.11 µs ┃ 408.5 % ┃ ┃ (example chain1/execute false 10) ┃ 1.13 ms ┃ ± 54.85 µs ┃ 428.5 % ┃ ┃ (example chain2/execute false 10) ┃ 1.86 ms ┃ ± 77.25 µs ┃ 704.8 % ┃ ┃ (example chain1/execute false 50) ┃ 958.99 µs ┃ ± 44.39 µs ┃ 364.3 % ┃ ┃ (example chain2/execute false 50) ┃ 2.10 ms ┃ ± 78.32 µs ┃ 797.6 % ┃ ┃ (example chain1/execute false 100) ┃ 1.85 ms ┃ ± 69.48 µs ┃ 701.7 % ┃ ┃ (example chain2/execute false 100) ┃ 960.76 µs ┃ ± 36.04 µs ┃ 365.0 % ┃ ┗━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┻━━━━━━━━━━━┻━━━━━━━━━━━━━┻━━━━━━━━━┛
This has got me down because how can that old ugly broken code be faster? ┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━━┳━━━━━━━━━┓ ┃ Expression ┃ Mean ┃ Var ┃ Ratio ┃ ┣━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╋━━━━━━━━━━━╋━━━━━━━━━━━━━╋━━━━━━━━━┫ ┃ (example chain1/execute true 10) ┃ 291.22 µs ┃ ± 4.23 µs ┃ 110.4 % ┃ ┃ (example chain2/execute true 10) ┃ 267.34 µs ┃ ± 7.82 µs ┃ 101.4 % ┃ ┃ (example chain1/execute true 50) ┃ 293.59 µs ┃ ± 18.86 µs ┃ 111.3 % ┃ ┃ (example chain2/execute true 50) ┃ 2.23 ms ┃ ± 109.84 µs ┃ 846.2 % ┃ (slowest) ┃ (example chain1/execute true 100) ┃ 263.75 µs ┃ ± 11.87 µs ┃ 100.0 % ┃ (fastest) ┃ (example chain2/execute true 100) ┃ 1.05 ms ┃ ± 21.40 µs ┃ 398.7 % ┃ ┃ (example chain1/execute false 10) ┃ 1.15 ms ┃ ± 49.73 µs ┃ 437.8 % ┃ ┃ (example chain2/execute false 10) ┃ 1.90 ms ┃ ± 62.41 µs ┃ 719.1 % ┃ ┃ (example chain1/execute false 50) ┃ 986.54 µs ┃ ± 21.83 µs ┃ 374.0 % ┃ ┃ (example chain2/execute false 50) ┃ 2.04 ms ┃ ± 45.58 µs ┃ 773.3 % ┃ ┃ (example chain1/execute false 100) ┃ 1.79 ms ┃ ± 29.31 µs ┃ 677.0 % ┃ ┃ (example chain2/execute false 100) ┃ 927.77 µs ┃ ± 18.47 µs ┃ 351.8 % ┃ ┗━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┻━━━━━━━━━━━┻━━━━━━━━━━━━━┻━━━━━━━━━┛
…ise. Odd! ┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━┓ ┃ Expression ┃ Mean ┃ Var ┃ Ratio ┃ ┣━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╋━━━━━━━━━━━╋━━━━━━━━━━━━╋━━━━━━━━━┫ ┃ (example chain1/execute true 10) ┃ 294.55 µs ┃ ± 11.79 µs ┃ 113.0 % ┃ ┃ (example chain2/execute true 10) ┃ 260.61 µs ┃ ± 9.16 µs ┃ 100.0 % ┃ (fastest) ┃ (example chain1/execute true 50) ┃ 290.55 µs ┃ ± 12.64 µs ┃ 111.5 % ┃ ┃ (example chain2/execute true 50) ┃ 2.14 ms ┃ ± 79.23 µs ┃ 819.3 % ┃ (slowest) ┃ (example chain1/execute true 100) ┃ 263.40 µs ┃ ± 10.36 µs ┃ 101.1 % ┃ ┃ (example chain2/execute true 100) ┃ 1.08 ms ┃ ± 40.20 µs ┃ 413.2 % ┃ ┃ (example chain1/execute false 10) ┃ 1.10 ms ┃ ± 45.12 µs ┃ 423.9 % ┃ ┃ (example chain2/execute false 10) ┃ 1.81 ms ┃ ± 68.63 µs ┃ 693.3 % ┃ ┃ (example chain1/execute false 50) ┃ 940.31 µs ┃ ± 24.03 µs ┃ 360.8 % ┃ ┃ (example chain2/execute false 50) ┃ 2.06 ms ┃ ± 84.38 µs ┃ 789.3 % ┃ ┃ (example chain1/execute false 100) ┃ 1.82 ms ┃ ± 79.02 µs ┃ 697.4 % ┃ ┃ (example chain2/execute false 100) ┃ 966.00 µs ┃ ± 32.83 µs ┃ 370.7 % ┃ ┗━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┻━━━━━━━━━━━┻━━━━━━━━━━━━┻━━━━━━━━━┛
…just happy things are relatively close. More accurate results, perhaps, with the very slow bench (rather than the default quick bench). ┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━┓ ┃ Expression ┃ Mean ┃ Var ┃ Ratio ┃ ┣━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╋━━━━━━━━━━━╋━━━━━━━━━━━━╋━━━━━━━━━━━┫ ┃ (example chain1/execute false 5) ┃ 173.56 µs ┃ ± 2.30 µs ┃ 100.6 % ┃ ┃ (example chain2/execute false 5) ┃ 1.76 ms ┃ ± 26.54 µs ┃ 1,022.5 % ┃ ┃ (example chain1/execute false 10) ┃ 172.60 µs ┃ ± 1.88 µs ┃ 100.0 % ┃ (fastest) ┃ (example chain2/execute false 10) ┃ 936.96 µs ┃ ± 42.44 µs ┃ 542.9 % ┃ ┃ (example chain1/execute false 50) ┃ 2.06 ms ┃ ± 33.00 µs ┃ 1,193.7 % ┃ ┃ (example chain2/execute false 50) ┃ 1.07 ms ┃ ± 14.47 µs ┃ 621.7 % ┃ ┃ (example chain1/execute false 100) ┃ 1.77 ms ┃ ± 26.09 µs ┃ 1,025.8 % ┃ ┃ (example chain2/execute false 100) ┃ 256.64 µs ┃ ± 3.49 µs ┃ 148.7 % ┃ ┃ (example chain1/execute true 5) ┃ 1.07 ms ┃ ± 15.58 µs ┃ 621.5 % ┃ ┃ (example chain2/execute true 5) ┃ 257.37 µs ┃ ± 3.40 µs ┃ 149.1 % ┃ ┃ (example chain1/execute true 10) ┃ 281.25 µs ┃ ± 4.07 µs ┃ 162.9 % ┃ ┃ (example chain2/execute true 10) ┃ 187.22 µs ┃ ± 2.95 µs ┃ 108.5 % ┃ ┃ (example chain1/execute true 50) ┃ 928.81 µs ┃ ± 14.83 µs ┃ 538.1 % ┃ ┃ (example chain2/execute true 50) ┃ 2.07 ms ┃ ± 35.27 µs ┃ 1,197.0 % ┃ (slowest) ┃ (example chain1/execute true 100) ┃ 284.05 µs ┃ ± 4.73 µs ┃ 164.6 % ┃ ┃ (example chain2/execute true 100) ┃ 191.76 µs ┃ ± 3.14 µs ┃ 111.1 % ┃ ┗━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┻━━━━━━━━━━━┻━━━━━━━━━━━━┻━━━━━━━━━━━┛
Also, inlined function calls for speed Now version 2 is nearly always faster than version 1
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
I thought I could make it much faster, but instead it's only marginally faster.