Francesco Nigro

hi @MM53 this looks similar to `maxAllowableDiffFromDBTime` in Classic ActiveMQ (see https://activemq.apache.org/pluggable-storage-lockers for more info): I agree that it could be useful, but dangerous as well, because if the difference...

> You said a difference above a few milliseconds could already be an issue. Then we probably have to change the code anyway as currently all milliseconds were removed before...

This one should be safe to be merged

@michaelandrepearce I was waiting the comment of @clebertsuconic ``` postponing after the next release... as discussed. ``` So don't know yet what are the replacement/proposal to replace it yet

These are my results by using a single single-threaded acceptor for both clients and replication (on the live broker) to fairly compare epoll vs io_uring under load. The test is...

> whats status on this one? Im keen to merge it, happy to help contribute any last bits like slight code re-org on if statements, and docs that are needed...

@jbertram > just a friendly reminder that it's been a week. This looks like a nice feature so it would be nice to get it out to the community if...

Not as good as I wished:  Consider that I've used [benchmark](https://github.com/open-telemetry/opentelemetry-java/blob/50408d499f85d5761d0a5ed9bf9d77d5ff01fff5/sdk/trace/src/jmh/java/io/opentelemetry/sdk/trace/export/BatchSpanProcessorBenchmark.java) setting mpsc array q of capacity == `spanCount * 5` (where 5 is the number of threads calling...

I've investigated with async-profiler what's going on here... mpsc array q: 412 samples with 100 Hz profiler -> ~42% CPU usage  mpsc xadd array q: 282 samples with 100...

@anuraaga Sadly not: it's something related to how caches works on machine; we just need to reduce the dependency from the size operation to make progresses...