jmh-basics: JMH tutorial module (dead-code elimination, Blackhole, constant folding, warmup/forks, state scopes, -prof gc)

Eight benchmark/demo classes, a deterministic test suite, and 12 captured output files generated by scripts/run-all.sh on JDK 25.0.5 (Temurin) with JMH 1.37. Co-Authored-By: Claude Sonnet 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_017ic4b6Fj2JNt7dRev4pZPe
2026-11-08 18:31:11 +00:00
parent f6237d4117
commit db129fb1ce
27 changed files with 2018 additions and 0 deletions
@@ -0,0 +1,97 @@
+# JMH version: 1.37
+# VM version: JDK 25.0.5, OpenJDK 64-Bit Server VM, 25.0.5+8-LTS
+# VM invoker: /tmp/tools/jdk-25.0.5+8/bin/java
+# Blackhole mode: compiler (auto-detected, use -Djmh.blackhole.autoDetect=false to disable)
+# Warmup: 3 iterations, 1 s each
+# Measurement: 5 iterations, 1 s each
+# Timeout: 10 min per iteration
+# Threads: 2 threads, will synchronize iterations
+# Benchmark mode: Average time, time/op
+# Benchmark: com.ankurm.jmhbasics.StateScopeBenchmark.perThreadCounter
+
+# Run progress: 0.00% complete, ETA 00:00:32
+# Fork: 1 of 2
+# Warmup Iteration   1: 7.678 ns/op
+# Warmup Iteration   2: 8.112 ns/op
+# Warmup Iteration   3: 7.597 ns/op
+Iteration   1: 7.543 ns/op
+Iteration   2: 7.587 ns/op
+Iteration   3: 7.824 ns/op
+Iteration   4: 8.115 ns/op
+Iteration   5: 8.116 ns/op
+
+# Run progress: 25.00% complete, ETA 00:00:24
+# Fork: 2 of 2
+# Warmup Iteration   1: 7.846 ns/op
+# Warmup Iteration   2: 7.687 ns/op
+# Warmup Iteration   3: 8.156 ns/op
+Iteration   1: 7.978 ns/op
+Iteration   2: 8.031 ns/op
+Iteration   3: 7.915 ns/op
+Iteration   4: 8.456 ns/op
+Iteration   5: 7.931 ns/op
+
+
+Result "com.ankurm.jmhbasics.StateScopeBenchmark.perThreadCounter":
+  7.950 ±(99.9%) 0.401 ns/op [Average]
+  (min, avg, max) = (7.543, 7.950, 8.456), stdev = 0.265
+  CI (99.9%): [7.549, 8.351] (assumes normal distribution)
+
+
+# JMH version: 1.37
+# VM version: JDK 25.0.5, OpenJDK 64-Bit Server VM, 25.0.5+8-LTS
+# VM invoker: /tmp/tools/jdk-25.0.5+8/bin/java
+# Blackhole mode: compiler (auto-detected, use -Djmh.blackhole.autoDetect=false to disable)
+# Warmup: 3 iterations, 1 s each
+# Measurement: 5 iterations, 1 s each
+# Timeout: 10 min per iteration
+# Threads: 2 threads, will synchronize iterations
+# Benchmark mode: Average time, time/op
+# Benchmark: com.ankurm.jmhbasics.StateScopeBenchmark.sharedCounter
+
+# Run progress: 50.00% complete, ETA 00:00:16
+# Fork: 1 of 2
+# Warmup Iteration   1: 28.549 ns/op
+# Warmup Iteration   2: 35.212 ns/op
+# Warmup Iteration   3: 35.105 ns/op
+Iteration   1: 32.801 ns/op
+Iteration   2: 35.661 ns/op
+Iteration   3: 33.767 ns/op
+Iteration   4: 32.006 ns/op
+Iteration   5: 35.276 ns/op
+
+# Run progress: 75.00% complete, ETA 00:00:08
+# Fork: 2 of 2
+# Warmup Iteration   1: 33.768 ns/op
+# Warmup Iteration   2: 35.397 ns/op
+# Warmup Iteration   3: 37.498 ns/op
+Iteration   1: 34.394 ns/op
+Iteration   2: 33.455 ns/op
+Iteration   3: 28.463 ns/op
+Iteration   4: 15.008 ns/op
+Iteration   5: 35.151 ns/op
+
+
+Result "com.ankurm.jmhbasics.StateScopeBenchmark.sharedCounter":
+  31.598 ±(99.9%) 9.365 ns/op [Average]
+  (min, avg, max) = (15.008, 31.598, 35.661), stdev = 6.194
+  CI (99.9%): [22.233, 40.963] (assumes normal distribution)
+
+
+# Run complete. Total time: 00:00:33
+
+REMEMBER: The numbers below are just data. To gain reusable insights, you need to follow up on
+why the numbers are the way they are. Use profilers (see -prof, -lprof), design factorial
+experiments, perform baseline and negative tests that provide experimental control, make sure
+the benchmarking environment is safe on JVM/OS/HW level, ask for reviews from the domain experts.
+Do not assume the numbers tell you what you want them to tell.
+
+NOTE: Current JVM experimentally supports Compiler Blackholes, and they are in use. Please exercise
+extra caution when trusting the results, look into the generated code to check the benchmark still
+works, and factor in a small probability of new VM bugs. Additionally, while comparisons between
+different JVMs are already problematic, the performance difference caused by different Blackhole
+modes can be very significant. Please make sure you use the consistent Blackhole mode for comparisons.
+
+Benchmark                             Mode  Cnt   Score   Error  Units
+StateScopeBenchmark.perThreadCounter  avgt   10   7.950 ± 0.401  ns/op
+StateScopeBenchmark.sharedCounter     avgt   10  31.598 ± 9.365  ns/op