feat(observability): spin up micro-benchmarking framework #2163
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
product-auto-label
bot
added
size: m
|
Pushing up more updates in a couple @alkatrivedi @surbhigarg92, but this PR is currently a draft. |
odeke-em
force-pushed
the
observability-benchmarking
branch
from
8edd742 to
f0eb7e5
Compare
product-auto-label
bot
added
size: l
odeke-em
force-pushed
the
observability-benchmarking
branch
4 times, most recently
from
d48777b to
90fb9d7
Compare
This requires you to add to your environment SPANNER_RUN_BENCHMARKS=true and then it can be run by npm run observability-test whose results look like: ```shell Benchmarking Total Runs: 10000 databaseRunSelect1AsyncAwait RAM Untraced(206.453kB) vs Traced (240.953kB): increase by (34.500kB) or 16.71% RAM Untraced(206.453kB) vs Traced+OTEL(243.391kB): increase by (36.938kB) or 17.89% Time Untraced(911.057µs) vs Traced (1.014ms): increase by (102.749µs) or 11.28% Time Untraced(911.057µs) vs Traced+OTEL(1.017ms): increase by (106.116µs) or 11.65% databaseRunSelect1Callback RAM Untraced(210.219kB) vs Traced (238.797kB): increase by (28.578kB) or 13.59% RAM Untraced(210.219kB) vs Traced+OTEL(242.914kB): increase by (32.695kB) or 15.55% Time Untraced(890.877µs) vs Traced (997.541µs): increase by (106.664µs) or 11.97% Time Untraced(890.877µs) vs Traced+OTEL(1.019ms): increase by (127.997µs) or 14.37% databaseRunTransactionAsyncTxRunUpdate RAM Untraced(308.898kB) vs Traced (325.313kB): increase by (16.414kB) or 5.31% RAM Untraced(308.898kB) vs Traced+OTEL(339.836kB): increase by (30.938kB) or 10.02% Time Untraced(1.510ms) vs Traced (1.652ms): increase by (141.330µs) or 9.36% Time Untraced(1.510ms) vs Traced+OTEL(1.668ms): increase by (157.466µs) or 10.43% databaseRunTransactionAsyncTxRun RAM Untraced(298.977kB) vs Traced (326.227kB): increase by (27.250kB) or 9.11% RAM Untraced(298.977kB) vs Traced+OTEL(315.164kB): increase by (16.188kB) or 5.41% Time Untraced(1.450ms) vs Traced (1.581ms): increase by (130.870µs) or 9.03% Time Untraced(1.450ms) vs Traced+OTEL(1.615ms): increase by (165.426µs) or 11.41% ```
odeke-em
changed the title
odeke-em
force-pushed
the
observability-benchmarking
branch
2 times, most recently
from
62ef9d8 to
2488550
Compare
… other code too
We now have more isolated and more accurate values
```shell
Benchmarking Database
Total Runs: 20000
Warm up runs: 2500
databaseRunSelect1AsyncAwait
RAM Untraced(209.906kB) vs Traced (224.453kB): increase by (14.547kB) or 6.93%
RAM Untraced(209.906kB) vs Traced+OTEL(243.211kB): increase by (33.305kB) or 15.87%
Time Untraced(901.139µs) vs Traced (918.845µs): increase by (17.706µs) or 1.96%
Time Untraced(901.139µs) vs Traced+OTEL(1.019ms): increase by (118.223µs) or 13.12%
databaseRunSelect1Callback
RAM Untraced(209.789kB) vs Traced (224.641kB): increase by (14.852kB) or 7.08%
RAM Untraced(209.789kB) vs Traced+OTEL(242.500kB): increase by (32.711kB) or 15.59%
Time Untraced(902.862µs) vs Traced (920.289µs): increase by (17.427µs) or 1.93%
Time Untraced(902.862µs) vs Traced+OTEL(1.027ms): increase by (124.280µs) or 13.77%
databaseRunTransactionAsyncTxRunUpdate
RAM Untraced(309.352kB) vs Traced (323.734kB): increase by (14.383kB) or 4.65%
RAM Untraced(309.352kB) vs Traced+OTEL(324.258kB): increase by (14.906kB) or 4.82%
Time Untraced(1.508ms) vs Traced (1.516ms): increase by (7.340µs) or 0.49%
Time Untraced(1.508ms) vs Traced+OTEL(1.679ms): increase by (170.989µs) or 11.34%
databaseRunTransactionAsyncTxRun
RAM Untraced(299.117kB) vs Traced (311.320kB): increase by (12.203kB) or 4.08%
RAM Untraced(299.117kB) vs Traced+OTEL(313.602kB): increase by (14.484kB) or 4.84%
Time Untraced(1.460ms) vs Traced (1.469ms): increase by (8.468µs) or 0.58%
Time Untraced(1.460ms) vs Traced+OTEL(1.679ms): increase by (219.068µs) or 15.00%
✔ Database runs compared (403854ms)
Benchmark getTracer
tracerProvider
Total Runs: 20000
Warm up runs: 2500
withNullTracerProviderUnsetGlobalTracerProvider
RAM (2.273kB)
TimeSpent (12.171µs)
withTracerDefinedTracerProvider
RAM (2.695kB)
TimeSpent (14.074µs)
withTracerRegisteredGlobally
RAM (2.727kB)
TimeSpent (14.413µs)
```
odeke-em
force-pushed
the
observability-benchmarking
branch
from
2488550 to
4d5c897
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This requires you to add to your environment
and then it can be run by npm run observability-test whose results look like:
startTrace against 3 different scenarios
Database runs against mockspanner
from a quiet Linux machine
$ uname -a Linux emmanuel-PowerEdge-T30 5.4.0-198-generic #218-Ubuntu SMP Fri Sep 27 20:18:53 UTC 2024 x86_64 x86_64 x86_64 GNU/Linux