Skip to content

Commit

Permalink
add benchmark results
Browse files Browse the repository at this point in the history
  • Loading branch information
@mladedav
mladedav committed Jun 16, 2024
1 parent 02af691 commit ed19dca
Show file tree
Hide file tree
Showing 3 changed files with 270 additions and 4 deletions.
4 changes: 4 additions & 0 deletions README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -122,6 +122,10 @@ fn main() {
}
```

## Benchmarks

This crate strives to be faster than `tracing_subscriber`. See [benchmarks](docs/benchmark.md).

## Supported Rust Versions

`json-subscriber` is built against the latest stable release. The minimum supported version is 1.65.
Expand Down
15 changes: 11 additions & 4 deletions benches/tracing_subscriber.rs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -6,19 +6,26 @@ mod support;
use support::MultithreadedBench;

fn mk_dispatch() -> tracing::Dispatch {
json_subscriber_dispatch()
// tracing_subscriber_dispatch()
#[cfg(not(bench_tracing_baseline))]
{
json_subscriber_dispatch()
}

#[cfg(bench_tracing_baseline)]
{
tracing_subscriber_dispatch()
}
}

#[cfg(not(bench_tracing_baseline))]
fn json_subscriber_dispatch() -> tracing::Dispatch {
let collector = json_subscriber::fmt::Subscriber::builder()
.with_writer(sink)
.finish();
tracing::Dispatch::new(collector)
}

// This can be used to set a baseline to compare this implementation with.
#[allow(dead_code)]
#[cfg(bench_tracing_baseline)]
fn tracing_subscriber_dispatch() -> tracing::Dispatch {
let collector = tracing_subscriber::fmt::Subscriber::builder()
.json()
Expand Down
255 changes: 255 additions & 0 deletions docs/benchmark.md
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,255 @@
## Comparison with `tracing-subscriber`

You can run this the suite using `tracing-subscriber` with:

```
RUSTFLAGS='--cfg bench_tracing_baseline' cargo bench -- --save-baseline tracing-subscriber
```

And then run the benchmark again using this library:

```
cargo bench -- --baseline tracing-subscriber
```

### Results

```
new_span/single_thread/1
time: [187.69 ns 188.37 ns 189.08 ns]
thrpt: [5.2888 Melem/s 5.3086 Melem/s 5.3279 Melem/s]
change:
time: [-52.385% -52.213% -52.015%] (p = 0.00 < 0.05)
thrpt: [+108.40% +109.26% +110.02%]
Performance has improved.
Found 2 outliers among 100 measurements (2.00%)
2 (2.00%) high mild
new_span/multithreaded/1
time: [5.0995 µs 5.1833 µs 5.2690 µs]
thrpt: [189.79 Kelem/s 192.93 Kelem/s 196.10 Kelem/s]
change:
time: [-78.716% -77.920% -77.127%] (p = 0.00 < 0.05)
thrpt: [+337.19% +352.89% +369.85%]
Performance has improved.
Found 4 outliers among 100 measurements (4.00%)
2 (2.00%) low mild
2 (2.00%) high mild
new_span/single_thread/10
time: [1.9265 µs 1.9310 µs 1.9357 µs]
thrpt: [5.1662 Melem/s 5.1788 Melem/s 5.1908 Melem/s]
change:
time: [-50.981% -50.829% -50.664%] (p = 0.00 < 0.05)
thrpt: [+102.69% +103.37% +104.00%]
Performance has improved.
Found 13 outliers among 100 measurements (13.00%)
9 (9.00%) high mild
4 (4.00%) high severe
new_span/multithreaded/10
time: [9.4159 µs 9.5547 µs 9.7191 µs]
thrpt: [1.0289 Melem/s 1.0466 Melem/s 1.0620 Melem/s]
change:
time: [-74.784% -73.499% -72.480%] (p = 0.00 < 0.05)
thrpt: [+263.37% +277.34% +296.57%]
Performance has improved.
Found 7 outliers among 100 measurements (7.00%)
1 (1.00%) low mild
4 (4.00%) high mild
2 (2.00%) high severe
new_span/single_thread/50
time: [10.237 µs 10.278 µs 10.323 µs]
thrpt: [4.8433 Melem/s 4.8649 Melem/s 4.8842 Melem/s]
change:
time: [-48.086% -47.936% -47.756%] (p = 0.00 < 0.05)
thrpt: [+91.410% +92.072% +92.624%]
Performance has improved.
Found 6 outliers among 100 measurements (6.00%)
6 (6.00%) high mild
new_span/multithreaded/50
time: [26.990 µs 27.319 µs 27.667 µs]
thrpt: [1.8072 Melem/s 1.8302 Melem/s 1.8526 Melem/s]
change:
time: [-74.114% -73.534% -72.993%] (p = 0.00 < 0.05)
thrpt: [+270.27% +277.84% +286.30%]
Performance has improved.
Found 6 outliers among 100 measurements (6.00%)
1 (1.00%) low mild
5 (5.00%) high mild
event/root/single_threaded/1
time: [497.48 ns 498.85 ns 500.33 ns]
thrpt: [1.9987 Melem/s 2.0046 Melem/s 2.0101 Melem/s]
change:
time: [-58.535% -58.270% -57.920%] (p = 0.00 < 0.05)
thrpt: [+137.64% +139.64% +141.17%]
Performance has improved.
Found 6 outliers among 100 measurements (6.00%)
4 (4.00%) high mild
2 (2.00%) high severe
event/root/multithreaded/1
time: [5.8425 µs 5.9341 µs 6.0291 µs]
thrpt: [165.86 Kelem/s 168.52 Kelem/s 171.16 Kelem/s]
change:
time: [-78.009% -76.339% -73.910%] (p = 0.00 < 0.05)
thrpt: [+283.28% +322.64% +354.73%]
Performance has improved.
Found 4 outliers among 100 measurements (4.00%)
2 (2.00%) low mild
1 (1.00%) high mild
1 (1.00%) high severe
event/unique_parent/single_threaded/1
time: [641.67 ns 643.06 ns 644.60 ns]
thrpt: [1.5513 Melem/s 1.5551 Melem/s 1.5584 Melem/s]
change:
time: [-75.884% -75.830% -75.770%] (p = 0.00 < 0.05)
thrpt: [+312.71% +313.74% +314.66%]
Performance has improved.
Found 6 outliers among 100 measurements (6.00%)
5 (5.00%) high mild
1 (1.00%) high severe
event/unique_parent/multithreaded/1
time: [7.1797 µs 7.2862 µs 7.4150 µs]
thrpt: [134.86 Kelem/s 137.25 Kelem/s 139.28 Kelem/s]
change:
time: [-78.334% -77.411% -76.535%] (p = 0.00 < 0.05)
thrpt: [+326.17% +342.70% +361.56%]
Performance has improved.
Found 4 outliers among 100 measurements (4.00%)
1 (1.00%) low mild
3 (3.00%) high mild
event/shared_parent/multithreaded/1
time: [7.4960 µs 7.6215 µs 7.7475 µs]
thrpt: [129.07 Kelem/s 131.21 Kelem/s 133.40 Kelem/s]
change:
time: [-77.724% -76.392% -75.098%] (p = 0.00 < 0.05)
thrpt: [+301.58% +323.59% +348.92%]
Performance has improved.
Found 1 outliers among 100 measurements (1.00%)
1 (1.00%) high mild
event/multi-parent/multithreaded/1
time: [10.216 µs 10.376 µs 10.548 µs]
thrpt: [94.801 Kelem/s 96.376 Kelem/s 97.885 Kelem/s]
change:
time: [-76.674% -75.840% -74.968%] (p = 0.00 < 0.05)
thrpt: [+299.49% +313.91% +328.71%]
Performance has improved.
Found 12 outliers among 100 measurements (12.00%)
2 (2.00%) low mild
6 (6.00%) high mild
4 (4.00%) high severe
event/root/single_threaded/10
time: [4.8850 µs 4.9149 µs 4.9525 µs]
thrpt: [2.0192 Melem/s 2.0346 Melem/s 2.0471 Melem/s]
change:
time: [-58.129% -57.648% -57.100%] (p = 0.00 < 0.05)
thrpt: [+133.10% +136.12% +138.83%]
Performance has improved.
event/root/multithreaded/10
time: [13.448 µs 13.855 µs 14.321 µs]
thrpt: [698.29 Kelem/s 721.77 Kelem/s 743.61 Kelem/s]
change:
time: [-72.857% -71.089% -69.055%] (p = 0.00 < 0.05)
thrpt: [+223.16% +245.89% +268.42%]
Performance has improved.
Found 5 outliers among 100 measurements (5.00%)
4 (4.00%) high mild
1 (1.00%) high severe
event/unique_parent/single_threaded/10
time: [6.6395 µs 6.6628 µs 6.6878 µs]
thrpt: [1.4953 Melem/s 1.5009 Melem/s 1.5061 Melem/s]
change:
time: [-75.174% -75.093% -75.019%] (p = 0.00 < 0.05)
thrpt: [+300.31% +301.49% +302.81%]
Performance has improved.
Found 6 outliers among 100 measurements (6.00%)
6 (6.00%) high mild
event/unique_parent/multithreaded/10
time: [14.951 µs 15.161 µs 15.379 µs]
thrpt: [650.25 Kelem/s 659.60 Kelem/s 668.86 Kelem/s]
change:
time: [-81.822% -81.345% -80.810%] (p = 0.00 < 0.05)
thrpt: [+421.11% +436.05% +450.12%]
Performance has improved.
Found 3 outliers among 100 measurements (3.00%)
2 (2.00%) high mild
1 (1.00%) high severe
event/shared_parent/multithreaded/10
time: [18.349 µs 18.607 µs 18.881 µs]
thrpt: [529.62 Kelem/s 537.44 Kelem/s 544.98 Kelem/s]
change:
time: [-80.082% -79.607% -79.118%] (p = 0.00 < 0.05)
thrpt: [+378.89% +390.36% +402.06%]
Performance has improved.
Found 6 outliers among 100 measurements (6.00%)
1 (1.00%) low mild
4 (4.00%) high mild
1 (1.00%) high severe
event/multi-parent/multithreaded/10
time: [33.293 µs 33.730 µs 34.199 µs]
thrpt: [292.40 Kelem/s 296.47 Kelem/s 300.36 Kelem/s]
change:
time: [-83.615% -83.094% -82.570%] (p = 0.00 < 0.05)
thrpt: [+473.72% +491.49% +510.33%]
Performance has improved.
Found 6 outliers among 100 measurements (6.00%)
5 (5.00%) high mild
1 (1.00%) high severe
event/root/single_threaded/50
time: [25.595 µs 25.671 µs 25.755 µs]
thrpt: [1.9414 Melem/s 1.9477 Melem/s 1.9535 Melem/s]
change:
time: [-57.180% -57.038% -56.893%] (p = 0.00 < 0.05)
thrpt: [+131.98% +132.76% +133.54%]
Performance has improved.
Found 1 outliers among 100 measurements (1.00%)
1 (1.00%) high mild
event/root/multithreaded/50
time: [37.453 µs 39.038 µs 40.700 µs]
thrpt: [1.2285 Melem/s 1.2808 Melem/s 1.3350 Melem/s]
change:
time: [-74.368% -73.402% -72.358%] (p = 0.00 < 0.05)
thrpt: [+261.77% +275.97% +290.13%]
Performance has improved.
Found 9 outliers among 100 measurements (9.00%)
7 (7.00%) high mild
2 (2.00%) high severe
event/unique_parent/single_threaded/50
time: [32.737 µs 32.868 µs 33.005 µs]
thrpt: [1.5149 Melem/s 1.5212 Melem/s 1.5273 Melem/s]
change:
time: [-75.443% -75.361% -75.273%] (p = 0.00 < 0.05)
thrpt: [+304.41% +305.86% +307.22%]
Performance has improved.
Found 6 outliers among 100 measurements (6.00%)
6 (6.00%) high mild
event/unique_parent/multithreaded/50
time: [47.636 µs 48.315 µs 49.058 µs]
thrpt: [1.0192 Melem/s 1.0349 Melem/s 1.0496 Melem/s]
change:
time: [-83.557% -82.962% -82.335%] (p = 0.00 < 0.05)
thrpt: [+466.11% +486.91% +508.18%]
Performance has improved.
Found 2 outliers among 100 measurements (2.00%)
1 (1.00%) low mild
1 (1.00%) high mild
event/shared_parent/multithreaded/50
time: [67.386 µs 68.607 µs 69.865 µs]
thrpt: [715.67 Kelem/s 728.79 Kelem/s 742.00 Kelem/s]
change:
time: [-82.195% -81.690% -81.210%] (p = 0.00 < 0.05)
thrpt: [+432.20% +446.14% +461.64%]
Performance has improved.
Found 4 outliers among 100 measurements (4.00%)
3 (3.00%) low mild
1 (1.00%) high mild
event/multi-parent/multithreaded/50
time: [268.57 µs 272.25 µs 275.93 µs]
thrpt: [181.21 Kelem/s 183.65 Kelem/s 186.17 Kelem/s]
change:
time: [-89.114% -88.884% -88.642%] (p = 0.00 < 0.05)
thrpt: [+780.41% +799.63% +818.58%]
Performance has improved.
Found 10 outliers among 100 measurements (10.00%)
3 (3.00%) low severe
3 (3.00%) low mild
4 (4.00%) high mild
```

0 comments on commit ed19dca

Please sign in to comment.