[floxy]

Your upper limit (1000000) needs to come from runtime (user input/file), otherwise you could create a C++ template program to compute the answer at compile time. Also, you might be interested in the Benchmarks Game:

https://benchmarksgame-team.pages.debian.net/benchmarksgame/...

...where it looks like GCC C++ beats GraalVM in all the tests.

[Osiris-Team]

Do they compare the second run of Java? Because using GraalVM native image does not really benefit from runtime info and optimization.

[igouy]

Various things.

3,000th run of Java HotSpot.

https://benchmarksgame-team.pages.debian.net/benchmarksgame/...

[floxy]

I know there was a lot of teeth gnashing in the past about HotSpot warmup, and I think it mostly boiled down to not much of a difference for most tests.

[igouy]

Fastest JMH after startup & warmup

https://benchmarksgame-team.pages.debian.net/benchmarksgame/...