Jens Axboe

@axboe

Linux kernel IO dabbler @ Facebook

Park City, Utah
Joined March 2009

Tweets

You blocked @axboe

Are you sure you want to view these Tweets? Viewing Tweets won't unblock @axboe

  1. Pinned Tweet
    6 Apr 2019

    Now that io_uring will be in the upcoming 5.1 kernel, I spent some time writing up how it came to be and how it works. More formats likely coming, and I expect a few revisions too. For now, consider this v1.0 of the doc:

    Show this thread
    Undo
  2. Jan 31
    Undo
  3. Jan 30

    Busy updating io_uring man pages. Making everything current, but also including kernel version availability of previous features. You can always find the up-to-date set in liburing - at least for now, at some point I'll try and push these to the normal man pages package.

    Undo
  4. Jan 29

    5.6 merge window changes for io_uring have now been merged. So many goodies in there, can't wait to see how 5.7 shapes up in the coming months.

    Undo
  5. Jan 29
    Undo
  6. Retweeted
    Jan 27

    [$] Some 5.5 kernel development statistics

    Undo
  7. Retweeted
    Jan 28

    Initial implementation of io_uring for seastar posted. Need to do more testing, polish some details, and find some annoying bugs. But looking good!

    Undo
  8. Retweeted
    Jan 24

    io_uring based transport layer for ASP․NET? Yes please! A very early prototype, rough edges, missing features but LOTs of fun! Feel free to leave a comment or even better, a commit :)

    Undo
  9. Jan 24

    Always appreciate when Jon writes up some of my work, he does it so much better than I can.

    Undo
  10. Retweeted
    Jan 24

    I'll be giving a short talk at FOSDEM this year in the room on modern systems engineering: * why io_uring will replace your IO stack * high concurrency DB techniques * bucking conv. wisdom on error handling If you're around come say hi :)

    Undo
  11. Retweeted
    Jan 24

    just FYI, io_uring has been so far an extreme pleasure to work with. Most of the time I am spending is on refactoring and on the fact that seastar will have to keep supporting the old aio api. If we were starting from scratch with io_uring today it'd be even easier

    Undo
  12. Retweeted
    Jan 22
    Undo
  13. Jan 23

    Was great to meet you and discuss io_uring and netty (and just tech in general too). Already sparked a small patch set:

    Undo
  14. Retweeted
    Jan 21
    Replying to

    The libunifex project by demonstrates using sender/receiver with io_uring.

    Undo
  15. Jan 21

    I've changed the liburing license from LGPL to dual licensed LGPL/MIT. This should make it easier for some folks to adopt.

    Undo
  16. Retweeted
    Jan 19

    vs : simple echo server. io_uring +99% performance, -45% cpu usage. Wow. . 🏆io_uring🏆. Try for yourself: io_uring: epoll:

    Undo
  17. Jan 15

    It's a great honor, hope I can live up to it! Below is how I see myself recruiting new speakers.

    Undo
  18. Retweeted

    Heya heya! 9th edition of Kernel Recipes will take place in Paris, 28 to 30th september 2020. Stay tuned!

    Undo
  19. Retweeted
    Jan 13

    Just cut rio 0.6.2 which automatically handles submission, removing another potential source of misuse for folks who want to take advantage of incredible io_uring perf in . io_uring is a goldrush for file and network IO. This changes everything.

    Undo
  20. Retweeted

    One of my very favorite graphs.

    Undo
  21. Retweeted
    Jan 6

    Just cut rio 0.5! It's a pure-rust misuse-resistant io_uring library. io_uring will change everything about IO. The goldrush has started but folks don't realize it yet. lifetimes+RAII tricks ensure safety. go measure the o_direct example vs what you currently use :]

    Undo

Loading seems to be taking a while.

Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.

    You may also like

    ·