Add JMH Dependencies
To run it we can use:
Run All Benchmark
./jmh.sh
Run a single benchmark
# All benchmarks within IndexingBenchmark
./jmh.sh IndexingBenchmark
# Only one benchmark within IndexingBenchmark
./jmh.sh IndexingBenchmark.measureIndexingAsKafkaSerializedDocument
Run API LOG benchmark
./jmh.sh IndexAPILog
NOTE: If you're running the benchmark locally make sure astra/benchmarks/api_logs.txt is present. Else you can run using java -Djmh.api.log.file=/path/to/file.txt -Xms4g -Xmx4g benchmarks/target/benchmarks.jar IndexAPILog
To run it from IntelliJ directly https://plugins.jetbrains.com/plugin/7529-jmh-java-microbenchmark-harness
To find all the JMH supported options use the -h flag
./jmh.sh -h
For example to run a benchmark with async profiler
./jmh.sh -prof "async:libPath=/Users/vthacker/async-profiler-2.0-macos-x64/build/libasyncProfiler.so;dir=jmh-output/async-profiler/;output=flamegraph;direction=forward"
Format Code Before Committing -
mvn fmt:format -f benchmarks/pom.xml