Ekump/test oom benchmarks issue 3

[ekump]

DO NOT MERGE

What does this PR do?

A brief description of the change being made with this pull request.

Motivation

What inspired you to submit this pull request?

Additional Notes

Anything else we should know when reviewing?

How to test the change?

Describe here in detail how the change can be validated.

[dd-octo-sts]

Artifact Size Benchmark Report

aarch64-alpine-linux-musl

Artifact	Baseline	Commit	Change
/aarch64-alpine-linux-musl/lib/libdatadog_profiling.a	85.91 MB	85.91 MB	0% (0 B) 👌
/aarch64-alpine-linux-musl/lib/libdatadog_profiling.so	7.88 MB	7.88 MB	0% (0 B) 👌

aarch64-unknown-linux-gnu

Artifact	Baseline	Commit	Change
/aarch64-unknown-linux-gnu/lib/libdatadog_profiling.a	97.11 MB	97.11 MB	0% (0 B) 👌
/aarch64-unknown-linux-gnu/lib/libdatadog_profiling.so	10.61 MB	10.61 MB	0% (0 B) 👌

libdatadog-x64-windows

Artifact	Baseline	Commit	Change
/libdatadog-x64-windows/debug/dynamic/datadog_profiling_ffi.dll	25.46 MB	25.46 MB	0% (0 B) 👌
/libdatadog-x64-windows/debug/dynamic/datadog_profiling_ffi.lib	88.44 KB	88.44 KB	0% (0 B) 👌
/libdatadog-x64-windows/debug/dynamic/datadog_profiling_ffi.pdb	184.60 MB	184.61 MB	+0% (+8.00 KB) 👌
/libdatadog-x64-windows/debug/static/datadog_profiling_ffi.lib	946.40 MB	946.40 MB	0% (0 B) 👌
/libdatadog-x64-windows/release/dynamic/datadog_profiling_ffi.dll	8.32 MB	8.32 MB	0% (0 B) 👌
/libdatadog-x64-windows/release/dynamic/datadog_profiling_ffi.lib	88.44 KB	88.44 KB	0% (0 B) 👌
/libdatadog-x64-windows/release/dynamic/datadog_profiling_ffi.pdb	24.62 MB	24.62 MB	0% (0 B) 👌
/libdatadog-x64-windows/release/static/datadog_profiling_ffi.lib	49.04 MB	49.04 MB	0% (0 B) 👌

libdatadog-x86-windows

Artifact	Baseline	Commit	Change
/libdatadog-x86-windows/debug/dynamic/datadog_profiling_ffi.dll	22.06 MB	22.06 MB	0% (0 B) 👌
/libdatadog-x86-windows/debug/dynamic/datadog_profiling_ffi.lib	89.82 KB	89.82 KB	0% (0 B) 👌
/libdatadog-x86-windows/debug/dynamic/datadog_profiling_ffi.pdb	188.62 MB	188.64 MB	+.01% (+24.00 KB) 🔍
/libdatadog-x86-windows/debug/static/datadog_profiling_ffi.lib	935.37 MB	935.37 MB	0% (0 B) 👌
/libdatadog-x86-windows/release/dynamic/datadog_profiling_ffi.dll	6.43 MB	6.43 MB	0% (0 B) 👌
/libdatadog-x86-windows/release/dynamic/datadog_profiling_ffi.lib	89.82 KB	89.82 KB	0% (0 B) 👌
/libdatadog-x86-windows/release/dynamic/datadog_profiling_ffi.pdb	26.43 MB	26.43 MB	0% (0 B) 👌
/libdatadog-x86-windows/release/static/datadog_profiling_ffi.lib	46.65 MB	46.65 MB	0% (0 B) 👌

x86_64-alpine-linux-musl

Artifact	Baseline	Commit	Change
/x86_64-alpine-linux-musl/lib/libdatadog_profiling.a	76.59 MB	76.59 MB	0% (0 B) 👌
/x86_64-alpine-linux-musl/lib/libdatadog_profiling.so	8.78 MB	8.78 MB	0% (0 B) 👌

x86_64-unknown-linux-gnu

Artifact	Baseline	Commit	Change
/x86_64-unknown-linux-gnu/lib/libdatadog_profiling.a	92.11 MB	92.11 MB	0% (0 B) 👌
/x86_64-unknown-linux-gnu/lib/libdatadog_profiling.so	10.69 MB	10.69 MB	0% (0 B) 👌

[pr-commenter]

Benchmarks

Comparison

Benchmark execution time: 2026-07-09 18:20:23

Comparing candidate commit 14ed300 in PR branch ekump/test-oom-benchmarks-issue-3 with baseline commit e026a3c in branch main.

Found 19 performance improvements and 29 performance regressions! Performance is the same for 100 metrics, 0 unstable metrics.

Explanation

This is an A/B test comparing a candidate commit's performance against that of a baseline commit. Performance changes are noted in the tables below as:

• 🟩 = significantly better candidate vs. baseline
• 🟥 = significantly worse candidate vs. baseline

We compute a confidence interval (CI) over the relative difference of means between metrics from the candidate and baseline commits, considering the baseline as the reference.

If the CI is entirely outside the configured SIGNIFICANT_IMPACT_THRESHOLD (or the deprecated UNCONFIDENCE_THRESHOLD), the change is considered significant.

Feel free to reach out to #apm-benchmarking-platform on Slack if you have any questions.

More details about the CI and significant changes

You can imagine this CI as a range of values that is likely to contain the true difference of means between the candidate and baseline commits.

CIs of the difference of means are often centered around 0%, because often changes are not that big:

---------------------------------(------|---^--------)-------------------------------->
                              -0.6%    0%  0.3%     +1.2%
                                 |          |        |
         lower bound of the CI --'          |        |
sample mean (center of the CI) -------------'        |
         upper bound of the CI ----------------------'

As described above, a change is considered significant if the CI is entirely outside the configured SIGNIFICANT_IMPACT_THRESHOLD (or the deprecated UNCONFIDENCE_THRESHOLD).

For instance, for an execution time metric, this confidence interval indicates a significantly worse performance:

----------------------------------------|---------|---(---------^---------)---------->
                                       0%        1%  1.3%      2.2%      3.1%
                                                  |   |         |         |
       significant impact threshold --------------'   |         |         |
                      lower bound of CI --------------'         |         |
       sample mean (center of the CI) --------------------------'         |
                      upper bound of CI ----------------------------------'

scenario:credit_card/is_card_number/ 378282246310005

• 🟩 execution_time [-5.760µs; -5.659µs] or [-7.804%; -7.667%]
• 🟩 throughput [+1126080.914op/s; +1145363.204op/s] or [+8.312%; +8.454%]

scenario:credit_card/is_card_number/378282246310005

• 🟩 execution_time [-5.375µs; -5.283µs] or [-7.640%; -7.509%]
• 🟩 throughput [+1155112.956op/s; +1174055.549op/s] or [+8.127%; +8.260%]

scenario:credit_card/is_card_number/37828224631000521389798

• 🟩 execution_time [-7.468µs; -7.443µs] or [-14.040%; -13.994%]
• 🟩 throughput [+3059492.002op/s; +3070419.092op/s] or [+16.273%; +16.331%]

scenario:credit_card/is_card_number/x371413321323331

• 🟥 execution_time [+397.885ns; +400.545ns] or [+6.173%; +6.215%]
• 🟥 throughput [-9079994.426op/s; -9019170.186op/s] or [-5.852%; -5.813%]

scenario:credit_card/is_card_number_no_luhn/ 378282246310005

• 🟩 execution_time [-5.506µs; -5.444µs] or [-9.338%; -9.234%]
• 🟩 throughput [+1727059.631op/s; +1744851.381op/s] or [+10.183%; +10.288%]

scenario:credit_card/is_card_number_no_luhn/378282246310005

• 🟩 execution_time [-5.439µs; -5.392µs] or [-9.767%; -9.682%]
• 🟩 throughput [+1926341.888op/s; +1941947.297op/s] or [+10.728%; +10.815%]

scenario:credit_card/is_card_number_no_luhn/37828224631000521389798

• 🟩 execution_time [-7.472µs; -7.445µs] or [-14.048%; -13.999%]
• 🟩 throughput [+3061059.061op/s; +3072536.328op/s] or [+16.280%; +16.341%]

scenario:credit_card/is_card_number_no_luhn/x371413321323331

• 🟥 execution_time [+399.039ns; +401.521ns] or [+6.192%; +6.230%]
• 🟥 throughput [-9102098.826op/s; -9045937.063op/s] or [-5.866%; -5.830%]

scenario:msgpack_decoder::v05/high_sharing/2000

• 🟩 execution_time [-95.944µs; -95.081µs] or [-5.823%; -5.770%]
• 🟩 throughput [+74361.236op/s; +74997.166op/s] or [+6.127%; +6.179%]

scenario:msgpack_decoder::v05/low_sharing/200

• 🟩 execution_time [-17.053µs; -16.352µs] or [-4.638%; -4.448%]
• 🟩 throughput [+25384.000op/s; +26350.890op/s] or [+4.666%; +4.844%]

scenario:otlp/e2e_json/1x1000

• 🟥 execution_time [+383.346µs; +384.781µs] or [+9.292%; +9.326%]

scenario:otlp/encode_json/1x1000

• 🟥 execution_time [+373.744µs; +374.262µs] or [+20.632%; +20.661%]

scenario:profile_serialize_compressed_pprof_timestamped_x1000

• 🟩 execution_time [-48.021µs; -47.227µs] or [-4.860%; -4.780%]

scenario:vec_map/as_deduped_map/already_deduped/16

• 🟥 execution_time [+6.030ns; +6.065ns] or [+23.801%; +23.939%]

scenario:vec_map/as_deduped_map/already_deduped/8

• 🟩 execution_time [-0.884ns; -0.856ns] or [-5.629%; -5.448%]

scenario:vec_map/contains_key/128

• 🟥 execution_time [+1.155µs; +1.165µs] or [+7.561%; +7.625%]
• 🟥 throughput [-593803.442op/s; -588873.532op/s] or [-7.086%; -7.027%]

scenario:vec_map/contains_key/16

• 🟥 execution_time [+254.886ns; +255.156ns] or [+102.296%; +102.405%]
• 🟥 throughput [-32511346.149op/s; -32450680.283op/s] or [-50.629%; -50.534%]

scenario:vec_map/contains_key/64

• 🟥 execution_time [+480.220ns; +483.371ns] or [+11.696%; +11.773%]
• 🟥 throughput [-1642392.092op/s; -1631727.925op/s] or [-10.536%; -10.468%]

scenario:vec_map/contains_key/8

• 🟥 execution_time [+137.229ns; +137.663ns] or [+189.117%; +189.716%]
• 🟥 throughput [-72223012.530op/s; -72086789.040op/s] or [-65.508%; -65.385%]

scenario:vec_map/get_hit/128

• 🟥 execution_time [+4.028µs; +4.036µs] or [+32.180%; +32.243%]
• 🟥 throughput [-2493817.476op/s; -2489084.478op/s] or [-24.387%; -24.340%]

scenario:vec_map/get_hit/16

• 🟥 execution_time [+36.967ns; +37.221ns] or [+17.356%; +17.475%]
• 🟥 throughput [-11183350.753op/s; -11102737.871op/s] or [-14.887%; -14.780%]

scenario:vec_map/get_hit/64

• 🟥 execution_time [+955.196ns; +960.517ns] or [+27.975%; +28.130%]
• 🟥 throughput [-4119787.056op/s; -4093200.009op/s] or [-21.979%; -21.837%]

scenario:vec_map/get_hit/8

• 🟥 execution_time [+4.914ns; +5.037ns] or [+8.002%; +8.203%]
• 🟥 throughput [-9899360.421op/s; -9639578.018op/s] or [-7.599%; -7.400%]

scenario:vec_map/get_miss/8

• 🟩 execution_time [-1.778ns; -1.747ns] or [-23.695%; -23.282%]

scenario:vec_map/iter/128

• 🟥 execution_time [+7.099ns; +7.178ns] or [+6.832%; +6.908%]
• 🟥 throughput [-79614141.903op/s; -78753388.056op/s] or [-6.463%; -6.393%]

scenario:vec_map/iter/16

• 🟥 execution_time [+0.550ns; +0.561ns] or [+4.191%; +4.276%]
• 🟥 throughput [-50071691.823op/s; -49056316.327op/s] or [-4.104%; -4.021%]

scenario:vec_map/iter/8

• 🟥 execution_time [+0.562ns; +0.569ns] or [+8.429%; +8.523%]
• 🟥 throughput [-94280826.178op/s; -93166751.987op/s] or [-7.861%; -7.768%]

Benchmark execution time: 2026-07-09 18:22:44

Comparing candidate commit 14ed300 in PR branch ekump/test-oom-benchmarks-issue-3 with baseline commit e026a3c in branch main.

Found 10 performance improvements and 6 performance regressions! Performance is the same for 107 metrics, 10 unstable metrics.

Explanation

This is an A/B test comparing a candidate commit's performance against that of a baseline commit. Performance changes are noted in the tables below as:

• 🟩 = significantly better candidate vs. baseline
• 🟥 = significantly worse candidate vs. baseline

We compute a confidence interval (CI) over the relative difference of means between metrics from the candidate and baseline commits, considering the baseline as the reference.

If the CI is entirely outside the configured SIGNIFICANT_IMPACT_THRESHOLD (or the deprecated UNCONFIDENCE_THRESHOLD), the change is considered significant.

Feel free to reach out to #apm-benchmarking-platform on Slack if you have any questions.

More details about the CI and significant changes

You can imagine this CI as a range of values that is likely to contain the true difference of means between the candidate and baseline commits.

CIs of the difference of means are often centered around 0%, because often changes are not that big:

---------------------------------(------|---^--------)-------------------------------->
                              -0.6%    0%  0.3%     +1.2%
                                 |          |        |
         lower bound of the CI --'          |        |
sample mean (center of the CI) -------------'        |
         upper bound of the CI ----------------------'

As described above, a change is considered significant if the CI is entirely outside the configured SIGNIFICANT_IMPACT_THRESHOLD (or the deprecated UNCONFIDENCE_THRESHOLD).

For instance, for an execution time metric, this confidence interval indicates a significantly worse performance:

----------------------------------------|---------|---(---------^---------)---------->
                                       0%        1%  1.3%      2.2%      3.1%
                                                  |   |         |         |
       significant impact threshold --------------'   |         |         |
                      lower bound of CI --------------'         |         |
       sample mean (center of the CI) --------------------------'         |
                      upper bound of CI ----------------------------------'

scenario:ddsketch_read/ordered_bins/clustered_near_zero

• 🟥 execution_time [+362.774ns; +365.725ns] or [+6.484%; +6.537%]

scenario:ddsketch_read/ordered_bins/collapsing

• 🟥 execution_time [+1.729µs; +1.741µs] or [+6.995%; +7.043%]

scenario:ddsketch_read/ordered_bins/large_values

• 🟥 execution_time [+469.476ns; +472.617ns] or [+6.573%; +6.617%]

scenario:normalization/normalize_name/normalize_name/Too-Long-.Too-Long-.Too-Long-.Too-Long-.Too-Long-.Too-Lo...

• 🟩 execution_time [-18.894µs; -18.714µs] or [-9.214%; -9.126%]
• 🟩 throughput [+490078.952op/s; +494484.674op/s] or [+10.050%; +10.140%]

scenario:normalization/normalize_name/normalize_name/bad-name

• 🟩 execution_time [-1.461µs; -1.413µs] or [-7.821%; -7.567%]
• 🟩 throughput [+4390003.415op/s; +4537239.340op/s] or [+8.200%; +8.475%]

scenario:normalization/normalize_name/normalize_name/good

• 🟩 execution_time [-1.074µs; -1.028µs] or [-9.895%; -9.472%]
• 🟩 throughput [+9667129.966op/s; +10063610.502op/s] or [+10.487%; +10.917%]

scenario:normalization/normalize_service/normalize_service/A0000000000000000000000000000000000000000000000000...

• 🟩 execution_time [-37.511µs; -37.089µs] or [-6.990%; -6.912%]
• 🟩 throughput [+138413.396op/s; +140020.388op/s] or [+7.428%; +7.514%]

scenario:normalization/normalize_service/normalize_service/Test Conversion 0f Weird !@#$%^&**() Characters

• 🟩 execution_time [-25.864µs; -25.760µs] or [-13.170%; -13.117%]
• 🟩 throughput [+768987.193op/s; +772034.455op/s] or [+15.102%; +15.162%]

scenario:normalization/normalize_service/normalize_service/test_ASCII

• 🟥 execution_time [+2.103µs; +2.161µs] or [+4.537%; +4.662%]
• 🟥 throughput [-960940.670op/s; -936001.832op/s] or [-4.454%; -4.339%]

scenario:receiver_entry_point/report/2644

• 🟥 execution_time [+176.520µs; +186.985µs] or [+5.251%; +5.562%]

Candidate

Omitted due to size.

Baseline

Omitted due to size.