Skip to content

benchmarks

Directory actions

More options

Directory actions

More options

Latest commit

 

History

History

benchmarks

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
src/test
src/test
 
 
Readme.md
Readme.md
 
 

Readme.md

Microbenchmarks

Benchmarks in this module are suitable for things like:

  • performance of SpEL evaluation
  • handling Context operations and probably not for more end to end scenarios - like testing Flink process

For some benchmarks sample values for 4-core machine are included. They should not be considered as proper reference, since results will vary a lot between machines/runs. However, if after some changes the results are like 10 times worse it's probably worth having a closer look ;)

Docs:

https://github.com/sbt/sbt-jmh

Running in sbt:

benchmarks/Jmh/run -i 8 -wi 3 -f1 -t2 .*SampleSpelBenchmark.*
benchmarks/Jmh/run -i 8 -wi 3 -f1 -t2 -prof jfr:stackDepth=512 .*SampleSpelBenchmark.*

E2E benchmarks

Benchmarks based on the docker installation example environment. The docker from sources is built and published in local registry and then the benchmark code is run.

Running in sbt:

bash -c "export NUSSKNACKER_SCALA_VERSION=2.13 && sbt \"benchmarks/test:runMain pl.touk.nussknacker.engine.benchmarks.e2e.FlinkSteamingScenarioBenchmark 10000000\""