LuciferYang commented on PR #48128:
URL: https://github.com/apache/spark/pull/48128#issuecomment-2354756826
> In addition, Apple Silicon benchmark result is quite misleading because
it's not the same Server-side result. That's one of the reasons why Apache
Spark community decided to run on GitHub Action. If we have a benchmark in
Spark codebase, why don't we regenerate the result in this PR, @George314159 ?
+1 for run the benchmark on GitHub Action.
Locally, I obtained the opposite result after switching the execution order
of `HashMap` and `LongMap`.
```scala
...
benchmark.addCase("LongMap") { _ =>
keys.foreach { key => map2.getOrElseUpdate(key, new Object()) }
}
benchmark.addCase("HashMap") { _ =>
keys.foreach { key => map1.getOrElseUpdate(key, new Object()) }
}
...
```
```
OpenJDK 64-Bit Server VM 17.0.12+7-LTS on Mac OS X 14.6.1
Apple M3
Benchmark Map: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
LongMap 443 450
8 0.0 110752.9 1.0X
HashMap 546 785
101 0.0 136431.3 0.8X
```
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
