> But my students weren’t as happy as I was - they wanted to build something genuinely useful, and they were really disappointed that our “product” had strong architectural limits and couldn’t outperform titans like nginx and haproxy.

I took a (very brief) look at the github repo [1], it doesn't look like you're doing anything with cpu pinning.

You can probably eke (thanks) out a bit more performance if you cpu pin your threads and cpu pin your listen sockets (sockopt SO_INCOMING_CPU).

If you also cpu align your outgoing sockets, you should get a significant boost, but afaik, there's no great api for that. Linux does have an api for compatible NICs (traffic steering/flow steering) which can work, but if you know what hash your NIC uses (it's probably toeplitz) and you manage source port selection to your backend, you can pick ports that will hash properly.

The goal is for your proxy to be able to handle packets without any cross cpu communication.

[1] https://github.com/sibexico/TinyGate

Take a look at https://github.com/concurrencykit/ck and https://github.com/microsoft/mimalloc, it will fit well for a zero-copy and mem aligned reverse proxy. Also, if you want to add a DDoS protection and more advanced L4 stuff check out https://docs.ebpf.io/ebpf-library/libxdp/libxdp/

Such a great article!

This sent me through a rabbit hole of uring, kernel development and C. I've been a rust and c++ dev for quite a few years now, but there's such a simplicity and even artistic feel to small(ish) C programs.

This article is very brief and not deep enough. There is one I’m finding a much more “to the bone” and explaining io_uring concepts in much deeper details:

https://unixism.net/loti/index.html

Feel free to reverse-search the related HN thread too.

Yes, io_uring is significantly faster than epoll (I think I had like 20% faster req/s with io_uring.) The catch is that its kernel opt-in and disabled just about everywhere for security reasons. I think that it has direct memory sharing between the kernel and user-land which is kind of yikes. There's been multiple exploits that hit io_uring in recent times. It's because of this that even engineering projects that try to reach the highest performance possible (like Go) don't really bake io_uring in as a sane default. Though if you want to take the risk you can always run it yourself for your favourite language. It is faster but the cost is possible exploits.

The main reason why it gets disabled is fixed now, the latest RC got cBPF support and as such you can restrict what OPs can be run now instead of just fully disabling it.

Tried the cBPF restricted ring in a 6.7-ish RC build last year, filtering to just READ/WRITE/OPENAT cut our attack surface enough that infra finally let us ship it prod-side. Gotcha: registering the filter wrong silently no-ops instead of erroring, took a day to notice ops just weren't landing.

Quite depends, I had times when my posix emulation of io_uring (with poll, not epoll) was faster than io_uring. For large zero-copy buffers, io_uring is king however. Also io_uring is useful even for non asynchronous IO as it can implement chain of operations as single atomic operation (mkdir + open it for example).

For something like networking, if you are maximizing packets per second, you'll hit kernel limits[1] very quickly and instead have to start leveraging features like GSO/GRO or completely bypass the network stack.

1: https://github.com/axboe/liburing/discussions/1346

RHEL 9 and 10 now fully support io_uring by default. It is very recent, but this covers a lot of corporate Linux installs. Gemini 'said' Ubuntu and SuSE support it as well, but did not provide any links to prove it.

https://access.redhat.com/solutions/4723221

Go should reconsider support. They should have a 'go' at it.

For a project like Go, wouldn't it be an option to do one-time iouring feature detection in the runtime startup? Exploits are an issue for the entire OS, not the program choosing to use iouring, yeah?

Any kind of poll mode networking:

Rdma, dpdk, io_uring it’s really kind of up to the user to do the memory isolation

In io_urings case tho, you can’t do much because the rings are in the kernel.

I’m hopeful though that with Llm things will get better.

But it’s just hard problem to solve . Very difficult to do in the kernel itself, and folks don’t really even understand tuning for it.

I've not yet tested the shared buffers for my io uring based web server, but that's because instead of reading from a file and writing, i send directly from a mmaped region.

But really, I want to sendfile with io_uring, but that's not supported yet.

My writeup, with extra buzzwords like Rust and kTLS: https://blog.habets.se/2025/04/io-uring-ktls-and-rust-for-ze...

It was on HN too: https://news.ycombinator.com/item?id=44980865

Yeah, splice through a pipe works but the pipe buffer roundtrip kills throughput for anything high volume, we benchmarked it against sendfile a while back and it was noticeably slower. Fine for correctness testing though, just don't expect it to hit line rate on a loaded box.

FYI you can use sendfile ish with uring, since splice(2) is implemented. Not as user friendly as sendfile, but should work fairly similarly.

On the Linux side sendfile is implemented via splice. So it is a more generic API that covers the sendfile case.

In the context of a proxy one should mention epoll_wait busy poll. I've recently dived into this when reviewing low-latency options, and found that it's almost possible to do user space busy polling just for simple sockets, no DPDK/VMA/io_uring needed, and Fastly contributed to this and uses it.

It's too low level, I cannot even tell that I understand everything, only the concept, so I will just share some links. It works only per NAPI epoll context, and one cannot easily control NAPI ID, but if an entire machine is dedicated for a proxy one can do a simple trick of assinging sockets by NAPI ID to dedicated pollers.

In my use case, it was not a proxy, but N socket polling on a machine that then processes received data. It does not look feasible for such case, maybe round-robin polling of NAPI contexts from a single thread may work. What I would really want to have one day from the kernel is that I can easily tell it: trust me, I will poll this single socket eventually, never ever use IRQ path for it.

Previous HN discussion of the kernel feature: https://news.ycombinator.com/item?id=43749271 Nice presentation by the Fastly contributor, with nice diagrams making the big picture much easier to understand: https://netdevconf.info/0x18/docs/netdev-0x18-paper10-talk-s... LWN articles: https://lwn.net/Articles/1008399/, https://lwn.net/Articles/997491/, https://lwn.net/Articles/959462/ Kernel docs: https://docs.kernel.org/networking/napi.html#irq-mitigation

If you write one with DPDK, it'll be infinitely more complex but you'll have the opportunity to blow away nginx in performance.

If you make one run on an FPGA you it'll be even more complex.

The lesson is that cutting through abstraction like a hot knife through butter is a necessary mindset for performance but also makes things more difficult. Sockets and thread-per-connection were good approaches when networks were very slow relative to CPUs, and they're still often the simplest approach today.

I was also always curious about this and recently wrote a few implementations of an http file server to teach myself the key differences.

https://theconsensus.dev/p/2026/05/18/serving-files-three-wa...

Boost asio if you love C++ and asynchronous networking.

I switched out asio's epoll backend for its io_uring in a database server and CPU utilization shot up. Probably depends on usage and the specifics of how it's integrated into the event code.

No async io framework exists which utilizes everything io_uring can, they are all build around the poll model. As such io_uring will always be worse than the poll like abstractions.

The two things that make io_uring fast are chaining of operations and zero syscall mode, the former would require that all async io frameworks/libs would need to be rewritten to make use of that and then all user facing apps would also need to be rewritten since all you’d get now are completions to operations instead of waiting if you can run a operation.

That’s paradoxically what you can expect on a busy server - your CPU can spend time doing work that would have been previously IO wait time. Of course, it could be a bug in the implementation where you’re spinning doing no work erroneously, but depends on the details.

In addition to the other discussion. It's important to measure outcomes and not just look at the cpu meter...

At the same load, how did latency look for A vs B.

What was throughput and latency at maximum load like for A vs B. For whichever one had the smaller max throughput, what did latency look like for the other option.

For bonus points while testing: is there another observable metric to indicate available capacity, if cpu % free is less useful.

Classic.

Know that the increase in CPU utilization may mean you've improved the performance of your "database server," because now your CPU cores are waiting less on IO. It also may not mean this, but just looking at htop won't tell you either way.

I’ve replaced Asio recently with stright epoll event loop and got about 16% RPS better. That is for resonably sized SQL server, so be careful with nice precanned libraries.

Boost is so inconvenient, they're huge dynamic libraries that are a pain to build and use. Even when I was already using CMake, getting Boost installed in a way where it could be discovered was super annoying. (I was on Mac, though)

Asio also comes in standalone form and both versions are header-only. Not necessarily directly related to your comment but adding it on, anyway.

Some (most?) Boost libraries are header-only. Including Boost.Asio nowadays.

You can statically link boost

We static linked boost at my last shop after burning a week fighting cmake and dynamic linking on three different OSes. Once it's static it's a non issue, build once and forget it. The setup pain is real but it's a one time tax, not ongoing.

The year is 2050; there are 20 different ways to poll a socket on Linux.

Yes, even for io_uring. io_uring singshot and then multishot to go even faster.

20 ways to poll and 20 ways for on-call to get paged at 3am when someone picks the wrong one under load. Doesn't matter how fast the syscall is if nobody on the team can debug it drunk-tired.

Just use AF_XDP

That only applies for network sockets, and requires taking responsibility for the entire network protocol stack. Not something that should be done without a very compelling reason, and not applicable at all in this case, where storage devices are involved.

You are correct and I misread the article while skimming it.

The author takes a very benchmark focus on this topic which only says part of the story particularly for complex systems. Noticed that there are a number of very similar interface that exist on other platform like windows long before io_uring, but that does make Linux’s I/O system worse or slow than these platforms. A fast server is likely fast in either multiplexing or async API if implemented correctly in almost all cases.

I'm now a Windows developer, mostly working with Linux and FreeBSD. Thx for the point, I'll look how it works in Windows systems.

Equivalent in Windows is Registered I/O (RIO) for sockets.

Windows network development is really, really different from Unixy stuff. But you might have fun :)

There is no benchmark in the post. There is analysis, discussion and code examples for epoll and io_uring usage.

The reactor/proactor split here isn't new — Doug Schmidt's ACE work in the 90s formalized this exact tradeoff for user-space frameworks. What's new is the kernel doing completion dispatch itself, which is where the batching gains come from. The security pushback seems more about attack surface than raw throughput doubts.