[erlang-questions] How much load can supervisors handle?

[Chris Hicks]

I thought I remember the question being asked but didn't see anything in my searches. Though I will admit I was pressed for time at work and probably didn't do as good a job at searching as I really should have. Thank you everyone for responding and especially for the report on your experiences with supervisors.
Chris
> From: gleber.p@REDACTED
> Date: Thu, 25 Oct 2012 12:38:34 +0200
> Subject: Re: [erlang-questions] How much load can supervisors handle?
> To: jesper.louis.andersen@REDACTED
> CC: silent_vendetta@REDACTED; erlang-questions@REDACTED
> 
> > On Oct 24, 2012, at 11:18 PM, Chris Hicks <silent_vendetta@REDACTED> wrote:
> >> Does anyone have any experience running one supervisor with a thousand,
> > ten-thousand, or more, workers under it with a high rate of churn? Would
> > a dynamically expanding tree of supervisors, which would obviously need to be
> > balanced, be a good solution for this?
> 
> Lately my experience with simple_one_for_one was more or less like this:
> - thousands work without issues
> - tens of thousands are OK, but the limit is near and it can add some latency
> - hundred of thousands is stalling supervisor for minutes and system
> stops handling any requests which need spawning processes under the
> supervisor
> 
> We solved this problem by "sharding" supervisors (spawn as many
> supervisors as there are schedulers) to ensure that all CPU cores are
> used.
> 
> Another question: why does supervisors not use ETS tables? I believe
> it should be faster than "sets" for handling many thousands of entries
> (I haven't done any benchmarks here).
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://erlang.org/pipermail/erlang-questions/attachments/20121025/ed2b5217/attachment.htm>