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(bin): panic when server terminates #1824

Merged
merged 1 commit into from
Apr 15, 2024

Conversation

mxinden
Copy link
Collaborator

@mxinden mxinden commented Apr 15, 2024

The neqo-bin benchmarks spin up a neqo-server and then run various different client scenarios against it. Once done, they send the neqo-server a signal to terminate.

In case the server terminates early (e.g. due to an error), surface it to the user.

The `neqo-bin` benchmarks spin up a `neqo-server` and then run various different
client scenarios against it. Once done, they send the `neqo-server` a signal to
terminate.

In case the server terminates early (e.g. due to an error), surface it to the
user.
@larseggert larseggert enabled auto-merge April 15, 2024 11:12
@larseggert larseggert added this pull request to the merge queue Apr 15, 2024
Copy link

codecov bot commented Apr 15, 2024

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 93.13%. Comparing base (cf00098) to head (ddb9d67).
Report is 1 commits behind head on main.

Additional details and impacted files
@@            Coverage Diff             @@
##             main    #1824      +/-   ##
==========================================
- Coverage   93.13%   93.13%   -0.01%     
==========================================
  Files         117      117              
  Lines       36356    36356              
==========================================
- Hits        33860    33859       -1     
- Misses       2496     2497       +1     

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

Merged via the queue into mozilla:main with commit 34d9603 Apr 15, 2024
15 checks passed
Copy link

Benchmark results

Performance differences relative to cf00098.

  • drain a timer quickly time: [309.56 ns 317.21 ns 324.18 ns]
    change: [-2.0431% +0.1831% +2.5576%] (p = 0.88 > 0.05)
    No change in performance detected.

  • coalesce_acked_from_zero 1+1 entries
    time: [194.40 ns 195.00 ns 195.59 ns]
    change: [-0.3970% +0.0523% +0.5076%] (p = 0.82 > 0.05)
    No change in performance detected.

  • coalesce_acked_from_zero 3+1 entries
    time: [239.51 ns 240.07 ns 240.69 ns]
    change: [-0.2625% +0.1987% +0.6440%] (p = 0.41 > 0.05)
    No change in performance detected.

  • coalesce_acked_from_zero 10+1 entries
    time: [238.54 ns 239.48 ns 240.64 ns]
    change: [-0.7531% +0.0195% +0.7830%] (p = 0.96 > 0.05)
    No change in performance detected.

  • coalesce_acked_from_zero 1000+1 entries
    time: [219.22 ns 226.76 ns 244.11 ns]
    change: [+0.0474% +3.0988% +8.4764%] (p = 0.21 > 0.05)
    No change in performance detected.

  • RxStreamOrderer::inbound_frame()
    time: [118.94 ms 119.10 ms 119.35 ms]
    change: [+0.1058% +0.3700% +0.6116%] (p = 0.00 < 0.05)
    Change within noise threshold.

  • transfer/Run multiple transfers with varying seeds
    time: [119.41 ms 119.70 ms 119.98 ms]
    thrpt: [33.338 MiB/s 33.418 MiB/s 33.499 MiB/s]
    change:
    time: [+0.1021% +0.4451% +0.7743%] (p = 0.01 < 0.05)
    thrpt: [-0.7684% -0.4431% -0.1020%]
    Change within noise threshold.

  • transfer/Run multiple transfers with the same seed
    time: [120.62 ms 120.82 ms 121.04 ms]
    thrpt: [33.048 MiB/s 33.107 MiB/s 33.162 MiB/s]
    change:
    time: [+0.7375% +0.9762% +1.2005%] (p = 0.00 < 0.05)
    thrpt: [-1.1863% -0.9668% -0.7321%]
    Change within noise threshold.

  • 1-conn/1-100mb-resp (aka. Download)/client
    time: [1.0993 s 1.1118 s 1.1256 s]
    thrpt: [88.839 MiB/s 89.943 MiB/s 90.964 MiB/s]
    change:
    time: [-2.1192% -0.9284% +0.2852%] (p = 0.19 > 0.05)
    thrpt: [-0.2844% +0.9371% +2.1651%]
    No change in performance detected.

  • 1-conn/10_000-parallel-1b-resp (aka. RPS)/client
    time: [424.25 ms 426.48 ms 428.73 ms]
    thrpt: [23.325 Kelem/s 23.448 Kelem/s 23.571 Kelem/s]
    change:
    time: [-1.4633% -0.7458% +0.0493%] (p = 0.06 > 0.05)
    thrpt: [-0.0493% +0.7514% +1.4850%]
    No change in performance detected.

  • 1-conn/1-1b-resp (aka. HPS)/client
    time: [50.185 ms 50.567 ms 50.945 ms]
    thrpt: [19.629 elem/s 19.776 elem/s 19.926 elem/s]
    change:
    time: [-1.0284% -0.0223% +1.0237%] (p = 0.97 > 0.05)
    thrpt: [-1.0133% +0.0223% +1.0391%]
    No change in performance detected.

Client/server transfer results

Transfer of 134217728 bytes over loopback.

Client Server CC Pacing Mean [ms] Min [ms] Max [ms] Relative
msquic msquic 424.0 ± 35.5 342.3 460.0 1.00
neqo msquic reno on 793.4 ± 43.7 744.1 898.4 1.00
neqo msquic reno 780.5 ± 23.5 753.8 820.9 1.00
neqo msquic cubic on 814.7 ± 131.2 703.9 1175.1 1.00
neqo msquic cubic 797.3 ± 64.2 753.9 968.3 1.00
msquic neqo reno on 4308.4 ± 141.6 4092.3 4534.5 1.00
msquic neqo reno 4412.7 ± 240.1 4116.8 4964.6 1.00
msquic neqo cubic on 4422.0 ± 160.0 4158.3 4669.4 1.00
msquic neqo cubic 4454.5 ± 167.8 4224.5 4713.4 1.00
neqo neqo reno on 3470.8 ± 150.3 3204.8 3721.8 1.00
neqo neqo reno 3554.3 ± 359.7 2712.8 4079.1 1.00
neqo neqo cubic on 4074.8 ± 449.4 3082.1 4431.8 1.00
neqo neqo cubic 4312.9 ± 135.6 4094.4 4474.3 1.00

⬇️ Download logs

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.

2 participants