Add performance comparison to snakeyaml-engine #217
Conversation
| @State(Scope.Benchmark) | ||
| @BenchmarkMode(Mode.AverageTime) | ||
| @OutputTimeUnit(BenchmarkTimeUnit.MILLISECONDS) | ||
| class SnakeyamlEngineJvmLoadingTimeBenchmark { |
There was a problem hiding this comment.
It should be displayed as a separate series on the JVM graph in https://krzema12.github.io/snakeyaml-engine-kmp-benchmarks/dev/bench/
There was a problem hiding this comment.
Yes, it will be. Each benchmark is displayed as a separate chart by default (you can modify the index.html in gh-pages branch if you need otherwise)
There was a problem hiding this comment.
you can modify the index.html in gh-pages branch if you need otherwise
I tried tinkering with JavaScript to have both JVM series on a single graph, but I lost the battle 😅 Maybe another time! This is the feature request on the GitHub action's side: benchmark-action/github-action-benchmark#175
|
|
||
| @Benchmark | ||
| fun loadsOpenAiSchema(): Map<*, *> { | ||
| return with(fileSystem()) { |
There was a problem hiding this comment.
question: why don't use just a standard java input stream instead of okio source which later converted to a stream? It is more fair in my opinion
There was a problem hiding this comment.
I wanted to keep the setup code as close to the KMP test as possible, to keep the overhead roughly the same to be able to compare both. But thanks for this comment, I'll give it more thought!
...otlin/it/krzeminski/snakeyaml/engine/kmp/benchmark/SnakeyamlEngineJvmLoadingTimeBenchmark.kt
Outdated
Show resolved
Hide resolved
This way we should have a reference point to the implementation where SnakeYAML Engine KMP was forked from.