1. 16
    1. 2

      I cared a lot about achieving the highest possible performance, so I decided to write my own abstraction layer for the lower-level io_uring interface.

      And then no mention of benchmarks or anything in the post?

      And when I go to the repo it does have a graph of benchmarks but the labels don’t make sense to me since they all talk about BPF and gnet (what is that?). What does BPF have to do with io_uring?

      Not trying to be overly critical, just a bit confused.

      1. 2

        The README has a performance section that suggest that it’s significantly faster than gnet and gnet appears to be pretty much on-par with the best results for that particular benchmark. This would be amazing if true.

        Not stated as such but the code implementing the server for the benchmark appears to be here.