On Thu, 26 Feb 2015 08:19:00 +0100 Aymeric Augustin <[email protected]> wrote:
> On 26 févr. 2015, at 00:00, Antoine Pitrou > <[email protected]> wrote: > > > On Wed, 25 Feb 2015 23:44:33 +0100 > > Ludovic Gasc <[email protected]> wrote: > >> > >> I've disabled keep_alive in api_hour, I quickly tested on agents list > >> webservices via localhost, I've 3334.52 req/s instead of 4179 req/s, 0.233 > >> latency average instead of 0.098 and 884 errors instead of 0 errors. > >> It isn't a big change compare to others Web frameworks values, but it's a > >> change. > > > > IMO, the fact that you get so many errors indicates that something is > > probably wrong in your benchmark setup. It is difficult to believe that > > Flask and Django would believe so badly in such a simple (almost > > simplistic) workload. > > If you push concurrency too far — like 5000 threads — I expect > performance to plummet and that kind of results. Shouldn't the server limit the size of the thread pool and queue incoming connections? That's what e.g. Apache will do. Regards Antoine.
