diff options
Diffstat (limited to '_build/static/archives/extend/2013-September/000228.html')
| -rw-r--r-- | _build/static/archives/extend/2013-September/000228.html | 133 |
1 files changed, 133 insertions, 0 deletions
diff --git a/_build/static/archives/extend/2013-September/000228.html b/_build/static/archives/extend/2013-September/000228.html new file mode 100644 index 00000000..88282495 --- /dev/null +++ b/_build/static/archives/extend/2013-September/000228.html @@ -0,0 +1,133 @@ +<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"> +<HTML> + <HEAD> + <TITLE> [99s-extend] Cowboy load test + </TITLE> + <LINK REL="Index" HREF="index.html" > + <LINK REL="made" HREF="mailto:extend%40lists.ninenines.eu?Subject=Re%3A%20%5B99s-extend%5D%20Cowboy%20load%20test&In-Reply-To=%3C5236044B.5040905%40ninenines.eu%3E"> + <META NAME="robots" CONTENT="index,nofollow"> + <style type="text/css"> + pre { + white-space: pre-wrap; /* css-2.1, curent FF, Opera, Safari */ + } + </style> + <META http-equiv="Content-Type" content="text/html; charset=us-ascii"> + <LINK REL="Previous" HREF="000227.html"> + <LINK REL="Next" HREF="000229.html"> + </HEAD> + <BODY BGCOLOR="#ffffff"> + <H1>[99s-extend] Cowboy load test</H1> + <B>Loïc Hoguin</B> + <A HREF="mailto:extend%40lists.ninenines.eu?Subject=Re%3A%20%5B99s-extend%5D%20Cowboy%20load%20test&In-Reply-To=%3C5236044B.5040905%40ninenines.eu%3E" + TITLE="[99s-extend] Cowboy load test">essen at ninenines.eu + </A><BR> + <I>Sun Sep 15 21:02:35 CEST 2013</I> + <P><UL> + <LI>Previous message: <A HREF="000227.html">[99s-extend] Cowboy load test +</A></li> + <LI>Next message: <A HREF="000229.html">[99s-extend] Cowboy load test +</A></li> + <LI> <B>Messages sorted by:</B> + <a href="date.html#228">[ date ]</a> + <a href="thread.html#228">[ thread ]</a> + <a href="subject.html#228">[ subject ]</a> + <a href="author.html#228">[ author ]</a> + </LI> + </UL> + <HR> +<!--beginarticle--> +<PRE>This sounds like you have too many timers, or something else is using +the CPU. 40k connections is literally nothing. Also make sure the +clients are on a separate VM/machine. + +Either way 99% CPU for 10k sounds high, I've had that with 0 CPU (though +on real hardware). + +Someone a couple years back got above 1 million after fixing 'too many +timers issues' but I don't think it was on a medium instance. + +On 09/15/2013 07:01 PM, Joe Freeman wrote: +><i> Hi, +</I>><i> +</I>><i> I've started work on a project using Clojure, but I was wondering +</I>><i> whether (and secretly hoping that) Erlang would be a better fit, so I've +</I>><i> been load testing a few web server frameworks. I'm particularly +</I>><i> interested in how the server can handle a large number of concurrent +</I>><i> WebSocket connections, and the test I've been running is similar to Eric +</I>><i> Moritz's [1]. +</I>><i> +</I>><i> I've setup a simple Cowboy 'echo' server running on an EC2 instance +</I>><i> (m1.medium, as in Eric's test) which could comfortably handle 10k +</I>><i> concurrent WebSocket requests (as in Eric's results), while echoing +</I>><i> about 200 messages/second. The CPU usage of the VM at this point is +</I>><i> about 99%, but the server continues to handle up to 40k concurrent +</I>><i> connections with a consistent average response time (<30ms). Pushing the +</I>><i> test beyond this number results in a spike in response times and lots of +</I>><i> connection timeouts. +</I>><i> +</I>><i> 40k connections seems pretty good, but when comparing this to the same +</I>><i> test against a couple of Clojure/JVM-based frameworks (specifically +</I>><i> Aleph/Netty and http-kit) I find I can get higher numbers of concurrent +</I>><i> connections with slightly better average response times (100k +</I>><i> connections, <10ms response time) using much less CPU (~20%). In fact, +</I>><i> memory seems to be the limiting factor. +</I>><i> +</I>><i> So I have two questions: +</I>><i> +</I>><i> 1) Should I be concerned about the CPU usage in the Erlang/Cowboy test? +</I>><i> I have limited experience with Erlang so far, but 100% CPU feels like a +</I>><i> bad thing. +</I>><i> +</I>><i> 2) Is there any way to increase the performance of the cowboy server? +</I>><i> Are there any Erlang VM parameters I can change? The fact that the +</I>><i> Clojure/JVM tests (on the same machine) have managed to get to 100k +</I>><i> connections suggests that the limitation isn't being imposed by the +</I>><i> operating system (I've applied changes various changes to sysctl and +</I>><i> ulimit). +</I>><i> +</I>><i> (Perhaps an echo server isn't the best way to compare HTTP servers, but +</I>><i> it feels like a good starting point.) +</I>><i> +</I>><i> Thanks for any help. +</I>><i> +</I>><i> [1] <A HREF="https://github.com/ericmoritz/wsdemo/blob/results-v1/results.md">https://github.com/ericmoritz/wsdemo/blob/results-v1/results.md</A> - +</I>><i> the GitHub repo actually contains code for an Aleph server, but results +</I>><i> from this aren't included in the summary here. +</I>><i> +</I>><i> +</I>><i> +</I>><i> _______________________________________________ +</I>><i> Extend mailing list +</I>><i> <A HREF="https://lists.ninenines.eu/listinfo/extend">Extend at lists.ninenines.eu</A> +</I>><i> <A HREF="http://lists.ninenines.eu:81/listinfo/extend">http://lists.ninenines.eu:81/listinfo/extend</A> +</I>><i> +</I> + +-- +Loïc Hoguin +Erlang Cowboy +Nine Nines +<A HREF="http://ninenines.eu">http://ninenines.eu</A> + +</PRE> + +<!--endarticle--> + <HR> + <P><UL> + <!--threads--> + <LI>Previous message: <A HREF="000227.html">[99s-extend] Cowboy load test +</A></li> + <LI>Next message: <A HREF="000229.html">[99s-extend] Cowboy load test +</A></li> + <LI> <B>Messages sorted by:</B> + <a href="date.html#228">[ date ]</a> + <a href="thread.html#228">[ thread ]</a> + <a href="subject.html#228">[ subject ]</a> + <a href="author.html#228">[ author ]</a> + </LI> + </UL> + +<hr> +<a href="https://lists.ninenines.eu/listinfo/extend">More information about the Extend +mailing list</a><br> +</body></html> |