Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

bench(transfer): run with and without pacing #2006

Merged
merged 3 commits into from
Jul 25, 2024

Conversation

mxinden
Copy link
Collaborator

@mxinden mxinden commented Jul 24, 2024

Might help debug #1998.

Does reproduce regression described in #1998.

Copy link

codecov bot commented Jul 24, 2024

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 94.99%. Comparing base (608c0a7) to head (c38fa44).

Additional details and impacted files
@@            Coverage Diff             @@
##             main    #2006      +/-   ##
==========================================
- Coverage   95.00%   94.99%   -0.01%     
==========================================
  Files         112      112              
  Lines       36371    36371              
==========================================
- Hits        34554    34551       -3     
- Misses       1817     1820       +3     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

Copy link

github-actions bot commented Jul 24, 2024

Failed Interop Tests

QUIC Interop Runner, client vs. server

neqo-latest as client

neqo-latest as server

All results

Succeeded Interop Tests

QUIC Interop Runner, client vs. server

neqo-latest as client

neqo-latest as server

Unsupported Interop Tests

QUIC Interop Runner, client vs. server

neqo-latest as client

neqo-latest as server

Copy link

github-actions bot commented Jul 24, 2024

Benchmark results

Performance differences relative to 608c0a7.

coalesce_acked_from_zero 1+1 entries: No change in performance detected.
       time:   [193.69 ns 194.13 ns 194.62 ns]
       change: [-0.1709% +0.3416% +0.7615%] (p = 0.16 > 0.05)
       No change
Found 7 outliers among 100 measurements (7.00%)
  4 (4.00%) high mild
  3 (3.00%) high severe
coalesce_acked_from_zero 3+1 entries: No change in performance detected.
       time:   [237.24 ns 237.86 ns 238.53 ns]
       change: [-0.9237% -0.2223% +0.3655%] (p = 0.53 > 0.05)
       No change
Found 16 outliers among 100 measurements (16.00%)
  1 (1.00%) low mild
  5 (5.00%) high mild
  10 (10.00%) high severe
coalesce_acked_from_zero 10+1 entries: No change in performance detected.
       time:   [237.56 ns 238.46 ns 239.48 ns]
       change: [-0.2078% +0.2500% +0.6814%] (p = 0.28 > 0.05)
       No change
Found 7 outliers among 100 measurements (7.00%)
  7 (7.00%) high severe
coalesce_acked_from_zero 1000+1 entries: No change in performance detected.
       time:   [217.01 ns 217.22 ns 217.46 ns]
       change: [-0.3565% +0.3775% +1.1441%] (p = 0.34 > 0.05)
       No change
Found 11 outliers among 100 measurements (11.00%)
  4 (4.00%) high mild
  7 (7.00%) high severe
RxStreamOrderer::inbound_frame(): Change within noise threshold.
       time:   [119.59 ms 119.73 ms 119.87 ms]
       change: [-0.8027% -0.5599% -0.3648%] (p = 0.00 < 0.05)
       Change within
transfer/pacing-false/varying-seeds:
       time:   [40.606 ms 42.665 ms 44.776 ms]
       thrpt:  [89.333 MiB/s 93.753 MiB/s 98.508 MiB/s]
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high mild
transfer/pacing-true/varying-seeds:
       time:   [53.904 ms 57.043 ms 60.251 ms]
       thrpt:  [66.389 MiB/s 70.123 MiB/s 74.206 MiB/s]
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high mild
transfer/pacing-false/same-seed:
       time:   [47.705 ms 49.184 ms 50.619 ms]
       thrpt:  [79.022 MiB/s 81.328 MiB/s 83.848 MiB/s]
Found 4 outliers among 100 measurements (4.00%)
  4 (4.00%) low mild
transfer/pacing-true/same-seed:
       time:   [65.243 ms 72.132 ms 78.858 ms]
       thrpt:  [50.724 MiB/s 55.454 MiB/s 61.309 MiB/s]
1-conn/1-100mb-resp (aka. Download)/client: No change in performance detected.
       time:   [289.33 ms 296.78 ms 304.06 ms]
       thrpt:  [328.88 MiB/s 336.95 MiB/s 345.62 MiB/s]
change:
       time:   [-2.8352% +0.4152% +3.7902%] (p = 0.82 > 0.05)
       thrpt:  [-3.6518% -0.4135% +2.9179%]
       No change
1-conn/10_000-parallel-1b-resp (aka. RPS)/client: Change within noise threshold.
       time:   [405.05 ms 408.35 ms 411.72 ms]
       thrpt:  [24.289 Kelem/s 24.489 Kelem/s 24.688 Kelem/s]
change:
       time:   [-2.9356% -1.8216% -0.7290%] (p = 0.00 < 0.05)
       thrpt:  [+0.7344% +1.8554% +3.0243%]
       Change within
1-conn/1-1b-resp (aka. HPS)/client: No change in performance detected.
       time:   [68.225 ms 68.596 ms 69.017 ms]
       thrpt:  [14.489  elem/s 14.578  elem/s 14.657  elem/s]
change:
       time:   [-0.6471% +0.1703% +1.0195%] (p = 0.69 > 0.05)
       thrpt:  [-1.0092% -0.1700% +0.6514%]
       No change
Found 10 outliers among 100 measurements (10.00%)
  1 (1.00%) low severe
  1 (1.00%) low mild
  8 (8.00%) high severe

Client/server transfer results

Transfer of 33554432 bytes over loopback.

Client Server CC Pacing Mean [ms] Min [ms] Max [ms] Relative
msquic msquic 211.3 ± 130.5 102.2 629.5 1.00
neqo msquic reno on 331.0 ± 87.5 262.0 473.6 1.00
neqo msquic reno 334.4 ± 81.4 262.3 485.7 1.00
neqo msquic cubic on 292.2 ± 71.7 253.1 494.6 1.00
neqo msquic cubic 287.6 ± 33.6 257.0 351.7 1.00
msquic neqo reno on 203.7 ± 77.5 129.6 408.5 1.00
msquic neqo reno 183.9 ± 71.3 121.1 399.3 1.00
msquic neqo cubic on 198.4 ± 72.6 142.9 341.5 1.00
msquic neqo cubic 269.6 ± 120.9 144.7 552.7 1.00
neqo neqo reno on 213.3 ± 82.2 166.2 448.6 1.00
neqo neqo reno 176.7 ± 20.5 144.7 227.5 1.00
neqo neqo cubic on 231.0 ± 91.4 169.0 424.6 1.00
neqo neqo cubic 232.6 ± 90.9 172.0 437.8 1.00

⬇️ Download logs

Copy link

github-actions bot commented Jul 24, 2024

Firefox builds for this PR

The following builds are available for testing. Crossed-out builds did not succeed.

Copy link
Collaborator

@larseggert larseggert left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We need a fix to the action/workflow that unbreaks the results reporting. (I'll see if I can come up with one.)

@larseggert larseggert added this pull request to the merge queue Jul 24, 2024
Merged via the queue into mozilla:main with commit aadf8be Jul 25, 2024
56 of 58 checks passed
@@ -200,7 +200,7 @@ jobs:
fi
sed -E -e 's/^ //gi' \
-e 's/((change|time|thrpt):[^%]*% )([^%]*%)(.*)/\1<b>\3<\/b>\4/gi' results.txt |\
perl -p -0777 -e 's/(.*?)\n(.*?)((No change|Change within|Performance has).*?)(\nFound .*?)?\n\n/<details><summary>$1: $3<\/summary><pre>\n$2$5<\/pre><\/details>\n/gs' |\
perl -p -0777 -e 's/(.*?)\n(.*?)(((No change|Change within|Performance has).*?)(\nFound .*?)?)?\n\n/<details><summary>$1: $4<\/summary><pre>\n$2$5$6<\/pre><\/details>\n/gs' |\
Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thank you @larseggert!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants