liburing VS Netty

I've tinkered around with io_uring on and off for the last couple years. But I think it's really becoming quite cool (not that it wasn't cool before... :)). This was a really interesting post on what's new https://github.com/axboe/liburing/wiki/io_uring-and-networki.... The combination of ring-mapped buffers and multi-shot operations has some really interesting applications for high-performance networking. Hoping over the next year or two we can start to see really bleeding edge networking perf without having to resort to using DPDK :)

Thought I was doing something wrong at first, but after looking at examples and code, I just wasn't able to reach the epoll numbers. Looking on the Github page, there a few issues there with people who found the same thing, with their own examples. #1, #2

To address my own silly questions, yes, one should use the new fixed buffers described in this document: https://github.com/axboe/liburing/wiki/io_uring-and-networki...

We're working on this! https://github.com/axboe/liburing/issues/830

Link: https://github.com/axboe/liburing/wiki/io_uring-and-networking-in-2023

Code of netty is here and using following library

We use Netty (https://netty.io/) as the source of the MQTT communication, and we build the MQTT features the MQTT broker should support ourselves on top of that.

In this space, we also have the somewhat related term blocking. Java's NIO library is one well-known non-blocking tool used for managing multiple tasks on a single Java thread. When listening to sockets, most of the time a thread is just blocked, doing nothing until it receives some data. So, it's efficient to use a single thread for monitoring many sockets, to increase the likelihood of the thread having some actual work to do. The Selector API does this but is notoriously challenging to program well. Instead, developers use frameworks like Netty which abstract some of NIO's complexity and layer on some best practices.

DirectByteBuffer exhibits an intriguing behavior: it deallocates its backing memory during the finalization process, which occurs after garbage collection (GC) cycles. This poses an issue if your system is conservative with on-heap allocations, leading to infrequent GC cycles. In such cases, there could be a significant delay between the time the memory becomes unreferenced and when it is actually deallocated. This behavior could, in some respects, mimic a memory leak.

This is why some libraries hacks into DirectByteBuffer to deallocate memory explicitly, bypassing the finalizer altogether. For instance, the Netty library has implemented such a workaround, see Netty as an [example](https://github.com/netty/netty/blob/795db4a866401aa172757b95...).

Apologies replying to myself, but Netty, which underpins many of the popular Java backend frameworks, see backward compatibility as more important than supporting green threads.

https://github.com/netty/netty/issues/12816

It'll be interesting to see who (if anyone) picks up Netty's mantle in the Project Loom world.

Looks fairly recent so I'm glad I had a pre oct22 build https://github.com/netty/netty/issues/12737

This is better? https://github.com/netty/netty/blob/4.1/pom.xml

Given the fact that Lettuce is built with Netty, we also immediately noticed quite an impact on the initialization time (cold start) of our lambda function. Netty is really fast while executing, but takes a bit of time to initialize. The new Lambda Snapstart functionality might help with that.