chore: prepare v0.13.0#2549
Conversation
Failed Interop TestsQUIC Interop Runner, client vs. server neqo-latest as client
neqo-latest as server
All resultsSucceeded Interop TestsQUIC Interop Runner, client vs. server neqo-latest as client
neqo-latest as server
Unsupported Interop TestsQUIC Interop Runner, client vs. server neqo-latest as client
neqo-latest as server
|
Benchmark resultsPerformance differences relative to d947c94. 1-conn/1-100mb-resp/mtu-1504 (aka. Download)/client: 💚 Performance has improved. time: [700.55 ms 705.48 ms 710.33 ms]
thrpt: [140.78 MiB/s 141.75 MiB/s 142.75 MiB/s]
change:
time: [-3.6796% -2.7335% -1.7463%] (p = 0.00 < 0.05)
thrpt: [+1.7773% +2.8103% +3.8202%]
1-conn/10_000-parallel-1b-resp/mtu-1504 (aka. RPS)/client: No change in performance detected. time: [348.48 ms 350.23 ms 351.97 ms]
thrpt: [28.412 Kelem/s 28.552 Kelem/s 28.696 Kelem/s]
change:
time: [-0.8573% -0.1160% +0.6344%] (p = 0.76 > 0.05)
thrpt: [-0.6304% +0.1161% +0.8647%]
1-conn/1-1b-resp/mtu-1504 (aka. HPS)/client: No change in performance detected. time: [25.686 ms 25.843 ms 25.998 ms]
thrpt: [38.465 elem/s 38.696 elem/s 38.932 elem/s]
change:
time: [-1.7789% -0.8794% +0.0043%] (p = 0.05 > 0.05)
thrpt: [-0.0043% +0.8872% +1.8111%]
1-conn/1-100mb-req/mtu-1504 (aka. Upload)/client: 💚 Performance has improved. time: [2.2949 s 2.3162 s 2.3376 s]
thrpt: [42.780 MiB/s 43.175 MiB/s 43.575 MiB/s]
change:
time: [-5.2546% -4.0278% -2.8255%] (p = 0.00 < 0.05)
thrpt: [+2.9076% +4.1968% +5.5461%]
decode 4096 bytes, mask ff: Change within noise threshold. time: [12.069 µs 12.099 µs 12.137 µs]
change: [+0.0380% +0.5587% +1.0865%] (p = 0.03 < 0.05)
decode 1048576 bytes, mask ff: 💔 Performance has regressed. time: [3.0747 ms 3.0836 ms 3.0935 ms]
change: [+3.8959% +4.3469% +4.8098%] (p = 0.00 < 0.05)
decode 4096 bytes, mask 7f: Change within noise threshold. time: [20.207 µs 20.270 µs 20.335 µs]
change: [+0.6421% +1.0114% +1.3731%] (p = 0.00 < 0.05)
decode 1048576 bytes, mask 7f: 💔 Performance has regressed. time: [5.2538 ms 5.2669 ms 5.2805 ms]
change: [+9.1750% +9.6070% +10.049%] (p = 0.00 < 0.05)
decode 4096 bytes, mask 3f: 💔 Performance has regressed. time: [7.0161 µs 7.0455 µs 7.0823 µs]
change: [+10.386% +11.250% +12.392%] (p = 0.00 < 0.05)
decode 1048576 bytes, mask 3f: 💚 Performance has improved. time: [1.7908 ms 1.7964 ms 1.8034 ms]
change: [-17.179% -16.767% -16.352%] (p = 0.00 < 0.05)
1 streams of 1 bytes/multistream: No change in performance detected. time: [71.474 µs 71.678 µs 71.886 µs]
change: [-2.3706% -0.3158% +1.1986%] (p = 0.78 > 0.05)
1000 streams of 1 bytes/multistream: Change within noise threshold. time: [25.359 ms 25.403 ms 25.448 ms]
change: [-0.4806% -0.2471% -0.0080%] (p = 0.04 < 0.05)
10000 streams of 1 bytes/multistream: No change in performance detected. time: [1.7010 s 1.7030 s 1.7050 s]
change: [-0.3082% -0.1366% +0.0223%] (p = 0.11 > 0.05)
1 streams of 1000 bytes/multistream: No change in performance detected. time: [73.075 µs 74.125 µs 75.617 µs]
change: [-2.6924% +0.1566% +3.0273%] (p = 0.92 > 0.05)
100 streams of 1000 bytes/multistream: Change within noise threshold. time: [3.4123 ms 3.4195 ms 3.4276 ms]
change: [+0.5495% +0.8292% +1.1091%] (p = 0.00 < 0.05)
1000 streams of 1000 bytes/multistream: Change within noise threshold. time: [145.92 ms 146.00 ms 146.09 ms]
change: [+0.0069% +0.0861% +0.1648%] (p = 0.03 < 0.05)
coalesce_acked_from_zero 1+1 entries: No change in performance detected. time: [94.731 ns 95.019 ns 95.310 ns]
change: [-0.7920% -0.1285% +0.5555%] (p = 0.73 > 0.05)
coalesce_acked_from_zero 3+1 entries: No change in performance detected. time: [112.52 ns 112.86 ns 113.22 ns]
change: [-0.6928% -0.2842% +0.1482%] (p = 0.21 > 0.05)
coalesce_acked_from_zero 10+1 entries: No change in performance detected. time: [111.74 ns 112.00 ns 112.34 ns]
change: [-0.8697% -0.2636% +0.4776%] (p = 0.44 > 0.05)
coalesce_acked_from_zero 1000+1 entries: No change in performance detected. time: [93.970 ns 94.380 ns 94.822 ns]
change: [-2.0716% -0.8335% +0.3671%] (p = 0.19 > 0.05)
RxStreamOrderer::inbound_frame(): 💚 Performance has improved. time: [114.61 ms 114.67 ms 114.73 ms]
change: [-2.2983% -2.2201% -2.1507%] (p = 0.00 < 0.05)
SentPackets::take_ranges: Change within noise threshold. time: [8.3622 µs 8.5685 µs 8.7607 µs]
change: [+0.9506% +4.3454% +7.9817%] (p = 0.01 < 0.05)
transfer/pacing-false/varying-seeds: Change within noise threshold. time: [34.509 ms 34.570 ms 34.630 ms]
change: [-2.8093% -2.5508% -2.3016%] (p = 0.00 < 0.05)
transfer/pacing-true/varying-seeds: Change within noise threshold. time: [34.612 ms 34.667 ms 34.722 ms]
change: [-2.1063% -1.8915% -1.6701%] (p = 0.00 < 0.05)
transfer/pacing-false/same-seed: Change within noise threshold. time: [34.330 ms 34.386 ms 34.443 ms]
change: [-2.9999% -2.7929% -2.5835%] (p = 0.00 < 0.05)
transfer/pacing-true/same-seed: Change within noise threshold. time: [35.039 ms 35.086 ms 35.132 ms]
change: [-3.1684% -2.9814% -2.7807%] (p = 0.00 < 0.05)
Client/server transfer resultsPerformance differences relative to d947c94. Transfer of 33554432 bytes over loopback, 30 runs. All unit-less numbers are in milliseconds.
|
|
Both merged. @larseggert fine if I proceed here? |
|
Go for it. |
|
For the record, downstream tracking issue: https://bugzilla.mozilla.org/show_bug.cgi?id=1959128 |
Bugfixes I would like to wait for:
Any objections? Anything else to include?