Hacker News new | past | comments | ask | show | jobs | submit login

NodeJS is insanely easy to do that with, it's one of the best reasons to use it in my experience:

    var queue = [];

    // request handling
    module.exports = function(request, response) {     
        queue.push("bla"); 
        response.end("bye"); 
    };

    // out-of-request work
    setInterval(function() { 
        // do something with your queue either locally
        // or moving it to an external job queue that can
        // leverage the same script on a different thread
    }, 1000);



what about concurrency?


It comes down to whether you're going to dominate node's thread or not, if it's heavy or of course blocking then you'll have to do push that work off to be handled by a different thread.

In my case I periodically query mongodb for new or updated data since the last time to be stored locally, push batches of data into redis, track and save load information etc all without any negative repercussions.


Too late to edit but I read you wrong lol. In terms of concurrency the out-of-request stuff is going to happen once per thread.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: