Francesco Nigro

ARTEMIS-3050 Reduce PagedReferenceImpl memory footprint

6

https://issues.apache.org/jira/browse/ARTEMIS-3050

ARTEMIS-3163 Experimental support for Netty IO_URING incubator

12

https://issues.apache.org/jira/browse/ARTEMIS-3163

(Further) reduce CPU usage of batch span processor

15

That's a follow-up issue of https://github.com/open-telemetry/opentelemetry-java/issues/2968 Same purpose, but as mentioned on https://github.com/open-telemetry/opentelemetry-java/issues/2968#issuecomment-864370074 I see that the referenced commit is using a CLQ with a separate size tracker to overcome...

Improve fairness of SimpleConnectionPool by yielding its combiner executor

6

Motivation: `SimpleConnectionPool` is backed by a `CombinerExecutor` that can be caught in a tight (potentially endless) loop while consuming (and executing) actions, occupying its current running event loop for too...

Support IO_URING transport

6

I'm implementing support for https://github.com/netty/netty-incubator-transport-io_uring on Vert-x `Transport`, see https://github.com/franz1981/vert.x/tree/4.2.1_iouring for some initial experiment.

Support IO_URING transport

14

Fixes #4163

Reduce InboundBuffer pending queue's allocation pressure

3

InboundBuffer's allocation create 16 pending entries by default while it could delay it when the initial size information is available. This change is going to lazy allocate it: it's less...

Event loop thread affinity

1

Allow users to define Netty event loop threads cores affinity. Given that Vert-x use long-living threads bound to specific selectors, being able to bind it to a specific core can...

HHH-15305 Update custom LIRS implementation based on BoundedConcurrentHashMap

1

This is using a separate size atomic counter. Potential improvements (need benchmarks to spot if makes sense to push this that far): - using separate queue add/remove counters using different...

Leak on server shutdown while still awaiting other nodes to join

9

Fixes #1096