Francesco Nigro

Done @mkouba and @geoand waiting the CI again, although I've just changed a comment and rebased)

@DeMol-EE I strongly suggest to extract GC logs or related JFR events too, to understand how the occupation of the heap as been - I suppose memory usage to be...

@geoand let me put it in draft, this is mostly to experiment, if you wanna give it a shot

I'm parking this here till we get some reliable estimation of overhead of the mentioned util methods

@johnaohara Did it happens if you switch off epoll native Netty and use just NIO?

And we can try with an explicit `-Dio.netty.noUnsafe=true` too, in case is due to something using 'Unsafe` that bother the GC

what's the status of this @vietj @MRuecker ?

yeah, at least `snappy` is very used (and I'm recently working with @lucamolteni to get it decent for us as well) and PRs are welcome as you might know ihih,...

Adding a comment here, as noted in the mentioned JEP there won't be any performance advantage (apart from saving inflation, which is still a concern!) If the Field/Methods are saved...

I'm banging my head how to reproduce it with simple actions, but failing ```yaml name: ordered-sequence phases: - single: always: users: 1 duration: 2 scenario: orderedSequences: - first: - addToInt:...