feat: Use isahc to replace hyper #471
Merged
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
Signed-off-by: Xuanwo <[email protected]>
Signed-off-by: Xuanwo <[email protected]>
Signed-off-by: Xuanwo <[email protected]>
Signed-off-by: Xuanwo <[email protected]>
Signed-off-by: Xuanwo <[email protected]>
Signed-off-by: Xuanwo <[email protected]>
|
isahc's maintainer confirms the benchmark. We will evaluate isahc for the next major release. |
Signed-off-by: Xuanwo <[email protected]>
Xuanwo
changed the title
Signed-off-by: Xuanwo <[email protected]>
Signed-off-by: Xuanwo <[email protected]>
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Outdated suggestions cannot be applied.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 hereby agree to the terms of the CLA available at: https://databend.rs/dev/policies/cla/
Summary
This PR is a demo of using isahc to replace hyper.
Conclusion:
ResponseFuturehas a lifetime. We have toBox::pinagain. (Can we improve?)Tested via local hosted minio on NVMe SSD:
isahc Bench:
Running benches/ops/main.rs (target/release/deps/ops-6744f7d05af32671) fs not set, ignore read_full/4.00 KiB time: [439.45 us 452.16 us 463.65 us] thrpt: [8.4250 MiB/s 8.6391 MiB/s 8.8889 MiB/s] Found 1 outliers among 100 measurements (1.00%) 1 (1.00%) high mild Benchmarking read_full/256 KiB: Collecting 100 samples in estimated 6.2662 s (10k iter read_full/256 KiB time: [522.37 us 551.01 us 578.93 us] thrpt: [431.83 MiB/s 453.71 MiB/s 478.59 MiB/s] Benchmarking read_full/4.00 MiB: Collecting 100 samples in estimated 5.0795 s (2200 it read_full/4.00 MiB time: [2.3599 ms 2.3756 ms 2.3911 ms] thrpt: [1.6336 GiB/s 1.6443 GiB/s 1.6552 GiB/s] Found 1 outliers among 100 measurements (1.00%) 1 (1.00%) high mild Benchmarking read_full/16.0 MiB: Collecting 100 samples in estimated 5.6395 s (700 ite read_full/16.0 MiB time: [8.0177 ms 8.0603 ms 8.1053 ms] thrpt: [1.9277 GiB/s 1.9385 GiB/s 1.9488 GiB/s] Found 38 outliers among 100 measurements (38.00%) 18 (18.00%) low mild 10 (10.00%) high mild 10 (10.00%) high severe Benchmarking read_part/4.00 KiB: Collecting 100 samples in estimated 5.3691 s (10k ite read_part/4.00 KiB time: [495.75 us 502.35 us 508.68 us] thrpt: [7.6792 MiB/s 7.7760 MiB/s 7.8795 MiB/s] Benchmarking read_part/256 KiB: Collecting 100 samples in estimated 6.4145 s (10k iter read_part/256 KiB time: [546.33 us 572.42 us 600.28 us] thrpt: [416.47 MiB/s 436.74 MiB/s 457.60 MiB/s] Found 9 outliers among 100 measurements (9.00%) 9 (9.00%) low mild Benchmarking read_part/4.00 MiB: Collecting 100 samples in estimated 5.0048 s (2100 it read_part/4.00 MiB time: [3.1035 ms 3.1304 ms 3.1580 ms] thrpt: [1.2369 GiB/s 1.2478 GiB/s 1.2586 GiB/s] Benchmarking read_part/16.0 MiB: Collecting 100 samples in estimated 5.8409 s (600 ite read_part/16.0 MiB time: [10.199 ms 10.379 ms 10.558 ms] thrpt: [1.4799 GiB/s 1.5055 GiB/s 1.5320 GiB/s] Benchmarking read_parallel/1x4.00 KiB: Collecting 100 samples in estimated 7.3937 s (1 read_parallel/1x4.00 KiB time: [466.93 us 475.74 us 484.27 us] thrpt: [8.0663 MiB/s 8.2109 MiB/s 8.3658 MiB/s] Found 4 outliers among 100 measurements (4.00%) 4 (4.00%) low mild Benchmarking read_parallel/2x4.00 KiB: Collecting 100 samples in estimated 5.1543 s (1 read_parallel/2x4.00 KiB time: [498.55 us 514.86 us 529.92 us] thrpt: [14.743 MiB/s 15.174 MiB/s 15.670 MiB/s] Benchmarking read_parallel/4x4.00 KiB: Collecting 100 samples in estimated 5.6615 s (1 read_parallel/4x4.00 KiB time: [524.40 us 537.37 us 550.92 us] thrpt: [28.362 MiB/s 29.077 MiB/s 29.796 MiB/s] Found 1 outliers among 100 measurements (1.00%) 1 (1.00%) high mild Benchmarking read_parallel/8x4.00 KiB: Collecting 100 samples in estimated 6.6902 s (1 read_parallel/8x4.00 KiB time: [645.51 us 658.73 us 672.32 us] thrpt: [46.481 MiB/s 47.440 MiB/s 48.411 MiB/s] Found 1 outliers among 100 measurements (1.00%) 1 (1.00%) high severe Benchmarking read_parallel/16x4.00 KiB: Warming up for 3.0000 s Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 5.2s, enable flat sampling, or reduce sample count to 60. Benchmarking read_parallel/16x4.00 KiB: Collecting 100 samples in estimated 5.1923 s ( read_parallel/16x4.00 KiB time: [958.84 us 981.99 us 1.0058 ms] thrpt: [62.138 MiB/s 63.646 MiB/s 65.183 MiB/s] Benchmarking read_parallel/1x256 KiB: Collecting 100 samples in estimated 7.0926 s (10 read_parallel/1x256 KiB time: [732.04 us 753.06 us 773.30 us] thrpt: [323.29 MiB/s 331.98 MiB/s 341.51 MiB/s] Found 8 outliers among 100 measurements (8.00%) 3 (3.00%) low mild 4 (4.00%) high mild 1 (1.00%) high severe Benchmarking read_parallel/2x256 KiB: Collecting 100 samples in estimated 7.5281 s (10 read_parallel/2x256 KiB time: [724.62 us 750.16 us 778.31 us] thrpt: [642.42 MiB/s 666.53 MiB/s 690.01 MiB/s] Benchmarking read_parallel/4x256 KiB: Collecting 100 samples in estimated 9.3891 s (10 read_parallel/4x256 KiB time: [875.97 us 895.56 us 916.19 us] thrpt: [1.0659 GiB/s 1.0904 GiB/s 1.1148 GiB/s] Found 1 outliers among 100 measurements (1.00%) 1 (1.00%) high mild Benchmarking read_parallel/8x256 KiB: Warming up for 3.0000 s Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 7.0s, enable flat sampling, or reduce sample count to 50. Benchmarking read_parallel/8x256 KiB: Collecting 100 samples in estimated 7.0137 s (50 read_parallel/8x256 KiB time: [1.3985 ms 1.4325 ms 1.4631 ms] thrpt: [1.3349 GiB/s 1.3634 GiB/s 1.3966 GiB/s] Benchmarking read_parallel/16x256 KiB: Collecting 100 samples in estimated 5.1749 s (2 read_parallel/16x256 KiB time: [2.1667 ms 2.2184 ms 2.2723 ms] thrpt: [1.7191 GiB/s 1.7608 GiB/s 1.8029 GiB/s] Benchmarking read_parallel/1x4.00 MiB: Collecting 100 samples in estimated 5.0887 s (1 read_parallel/1x4.00 MiB time: [3.1645 ms 3.2070 ms 3.2519 ms] thrpt: [1.2012 GiB/s 1.2180 GiB/s 1.2344 GiB/s] Found 3 outliers among 100 measurements (3.00%) 3 (3.00%) high mild Benchmarking read_parallel/2x4.00 MiB: Collecting 100 samples in estimated 5.4133 s (1 read_parallel/2x4.00 MiB time: [4.8317 ms 4.8898 ms 4.9526 ms] thrpt: [1.5775 GiB/s 1.5977 GiB/s 1.6169 GiB/s] Found 8 outliers among 100 measurements (8.00%) 8 (8.00%) high mild Benchmarking read_parallel/4x4.00 MiB: Collecting 100 samples in estimated 5.2386 s (5 read_parallel/4x4.00 MiB time: [9.9052 ms 9.9783 ms 10.050 ms] thrpt: [1.5547 GiB/s 1.5659 GiB/s 1.5775 GiB/s] Found 5 outliers among 100 measurements (5.00%) 4 (4.00%) low mild 1 (1.00%) high mild Benchmarking read_parallel/8x4.00 MiB: Collecting 100 samples in estimated 6.1177 s (2 read_parallel/8x4.00 MiB time: [29.730 ms 29.853 ms 29.980 ms] thrpt: [1.0424 GiB/s 1.0468 GiB/s 1.0511 GiB/s] Benchmarking read_parallel/16x4.00 MiB: Collecting 100 samples in estimated 7.1522 s ( read_parallel/16x4.00 MiB time: [29.877 ms 30.356 ms 30.851 ms] thrpt: [2.0258 GiB/s 2.0589 GiB/s 2.0919 GiB/s] Benchmarking read_parallel/1x16.0 MiB: Collecting 100 samples in estimated 5.2187 s (7 read_parallel/1x16.0 MiB time: [7.8764 ms 7.9292 ms 7.9798 ms] thrpt: [1.9581 GiB/s 1.9706 GiB/s 1.9838 GiB/s] Found 12 outliers among 100 measurements (12.00%) 12 (12.00%) low mild Benchmarking read_parallel/2x16.0 MiB: Collecting 100 samples in estimated 6.4420 s (4 read_parallel/2x16.0 MiB time: [15.509 ms 15.573 ms 15.653 ms] thrpt: [1.9964 GiB/s 2.0067 GiB/s 2.0149 GiB/s] Found 3 outliers among 100 measurements (3.00%) 2 (2.00%) high mild 1 (1.00%) high severe Benchmarking read_parallel/4x16.0 MiB: Collecting 100 samples in estimated 6.7507 s (2 read_parallel/4x16.0 MiB time: [31.874 ms 32.127 ms 32.371 ms] thrpt: [1.9307 GiB/s 1.9454 GiB/s 1.9608 GiB/s] Found 9 outliers among 100 measurements (9.00%) 1 (1.00%) low severe 7 (7.00%) low mild 1 (1.00%) high mild Benchmarking read_parallel/8x16.0 MiB: Warming up for 3.0000 s Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 7.2s, or reduce sample count to 60. Benchmarking read_parallel/8x16.0 MiB: Collecting 100 samples in estimated 7.1595 s (1 read_parallel/8x16.0 MiB time: [71.410 ms 72.214 ms 73.032 ms] thrpt: [1.7116 GiB/s 1.7310 GiB/s 1.7504 GiB/s] Benchmarking read_parallel/16x16.0 MiB: Warming up for 3.0000 s Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 15.1s, or reduce sample count to 30. Benchmarking read_parallel/16x16.0 MiB: Collecting 100 samples in estimated 15.051 s ( read_parallel/16x16.0 MiB time: [149.86 ms 150.95 ms 152.00 ms] thrpt: [1.6447 GiB/s 1.6561 GiB/s 1.6682 GiB/s]hyper Bench:
Also reported to sagebind/isahc#401