Live data from Hacker News

Viewing profile — bnoordhuis

bnoordhuis

HN member
Joined
Fri, Mar 19, 2010, 3:45 PM UTC
HN karma
604
Public activity
103 items

About bnoordhuis

Dutch hacker from the greater Rotterdam area. Always up for a beer. Contact me at info bnoordhuis nl.

https://github.com/bnoordhuis

Recent public activity

  1. comment
    Comment #9740250

    It's in the works: https://code.google.com/p/v8/issues/detail?id=811 You can tell by the number of CLs that a lot of work has been done already but there are a few more still in pr…

  2. comment
    Comment #9530625

    I've been going back and forth on this. On the one hand, I feel that positive discrimination is just as bad as the other kind and that its proponents are not thinking things throug…

  3. comment
    Comment #8911629

    Interesting results, thanks for sharing. I can perhaps shed some light on the performance differences. > Buffer 4.259 5.006 In v0.10, buffers are sliced off from big chunks of pre-…

  4. story
  5. comment
    Comment #6598527

    Libuv author here. Libuv doesn't try to be all things to all people - its main users are Node.js and Rust - but if you have suggestions on how to improve the API or the implementat…

  6. comment
    Comment #3431529

    On and off kernel hacker here. Compile and test the release candidates Linus puts out every fortnight. When the inevitable breakage happens, use `git bisect` to track down the offe…

  7. comment
    Comment #3199358

    That's a regression we'll probably have fixed in 0.6.1. Note that I/O read performance is up 35%, from 19.7 mb/s in 0.4 to 26.6 mb/s in 0.6.

  8. comment
    Comment #3190752

    I suspect that it's a jab at ALSA which, for reasons unknown, ships with a full LISP interpreter[1] that it doesn't actually use... Lennart Poetering is the author of PulseAudio. H…

  9. comment
    Comment #2891562

    > Isn't Node.js single threaded ? Yes (for now, we might take up V8 isolates). > Would it not under-perform , say compared to Erlang or Netty, in a multi-core CPU. No. You can spin…

  10. comment
    Comment #2786248

    It's been fixed in master. 0.5.2 comes out tomorrow (if all goes well) and will also contain the fix. https://github.com/joyent/libuv/issues/112 for details.

  11. comment
    Comment #2563683

    linux doesn't have kqueue. But kqueue is only an API, you could emulate it with io_submit() and io_getevents(). A bigger obstacle is that not all file systems support asynchronous …

  12. comment
    Comment #2558246

    That was the impression I walked away with too: much CS ivory tower thinking, little real-world relevance. But they're only interviews, everyday life at Google is probably (hopeful…

  13. comment
    Comment #2549813

    There isn't a shred of the original mongrel parser left. I'm not 100% sure but http-parser looks more like it's inspired by nginx than that it actually contains nginx code. Disclos…

  14. comment
    Comment #2548592

    Maybe I can help. I'm relatively local (Gouda, ZH) and I've managed teams like the one you mention before. Shoot me an email, address is in my profile.

  15. comment
    Comment #2514135

    The language was C and my solution (adapted to the article) was something like this: // TODO enforce! see also: soylent green #define MAX_AGE 100 struct student { const char *name;…

  16. comment
    Comment #2508576

    > His grandmother knew radix sort. Radix sort? I read it as a trick question where the answer is 'hash table with numberOfYearsOld as the key'. I suggested that as the solution to …

  17. comment
    Comment #2473742

    Sounds about right. My base salary, no benefits, would be about 58K a year if I worked full-time. This is in Rotterdam, where salaries are slightly lower than in Amsterdam.

  18. comment
    Comment #2470534

    I hear you, threading in SM is indeed fraught with dangers. On the off chance that you're going to revisit it, the way to go is to compile with -DJS_THREADSAFE and register a callb…

  19. comment
    Comment #2469945

    V8 is a single-threaded VM and I don't see that changing any time soon, it's very much married to Chrome's per-process model. SpiderMonkey doesn't have that drawback, it supports t…

  20. comment
    Comment #2468786

    You'll be hard-pressed to exhaust all ports: modern operating systems track connections by source address + source port + target address + target port. I wouldn't be surprised if t…

  21. comment
    Comment #2468345

    Take the staged approach: 1. Begin with a simple JDBC-backed servlet + JSP web app, like a blog (boring, I now), and deploy it to a servlet container like Jetty or Tomcat. 2. Migra…

  22. comment
    Comment #2396678

    Sounds promising. Is remote an option?

  23. comment
    Comment #2387577

    > Don't things like real time linux or soft real time patches for linux help, though ? I don't see how it could. The CPU essentially drops everything it's doing when a packet comes…

  24. comment
    Comment #2387202

    The problem is that a high-bandwidth link overruns the CPU with thousands of interrupts per second. It can paralyse the system, where the CPU is so busy handling I/O that it doesn'…

  25. comment
    Comment #2302042

    No. The main problem the event MPM addresses is HTTP keep-alive. With pre-fork and threaded, every connection takes up a process or thread. This sucks for keep-alive because it can…