Francesco Nigro

This is the improvement related: `master` ``` Benchmark Mode Cnt Score Error Units SymbolsBenchmark.decode avgt 5 283.784 ± 4.618 ns/op ``` ``` Benchmark Mode Cnt Score Error Units CompositeReadableBufferBenchmark.hashCodeMultipleArrayFullSpan avgt...

The changes on this PR (especially those that save creating a `slice` while decoding should benefit decoding in general :) I'm now testing this with and end 2 end test...

I've noticed a thing, due to the changes on decoding ie not creating the slice anymore: I'm getting a huge amount of `StringType$1.decode` unseen before, see  in violet

@gemmellr I've tried an end 2 end test but I'm still struggling to evaluate the perf improvement (that's strange, but not unexpected, according to our chats). I'm investigating what's happening...

@willr3 @jabolina qq for you folks: I see that @rvansa has enabled `preferIPv4Stack=true` pretty much everywhere to save the clustering to break: any hint why this could cause any harm?...

Ok so it's time to let this go: many thanks @wburns tomorrow I'll send a PR to remove it, and see if your clustering test will survive :P

many thanks @belaban and happy to hear from you, bud :)

See https://github.com/Hyperfoil/Hyperfoil/blob/5dbd941d579c47c1f5a03021468f266934ba9e80/core/src/main/java/io/hyperfoil/core/impl/PhaseInstanceImpl.java#L350 and the `nextDelta`, still in ms is at https://github.com/Hyperfoil/Hyperfoil/blob/5dbd941d579c47c1f5a03021468f266934ba9e80/core/src/main/java/io/hyperfoil/core/impl/PhaseInstanceImpl.java#L374 while the next fire time computation happen at https://github.com/Hyperfoil/Hyperfoil/blob/5dbd941d579c47c1f5a03021468f266934ba9e80/core/src/main/java/io/hyperfoil/core/impl/PhaseInstanceImpl.java#L380 which, in case the rate is pretty high (> 1000...

It would be nice to make uses of a GC-free microseconds time from epoch timer, if possible, too, without using JNI/JNA (and forcing us to have some specific platforms support)....

To be fair the name of this issue should be changed into: "bursty session's start while going past 1000 rps". In short, given that hyperfoil is using `double` fire time...