[erlang-questions] Any more recent version of Joe's "PerformanceMeasurements of Threads in Java and Processes in Erlang"

Jonathan Gold <>
Mon Apr 23 17:56:37 CEST 2007


Thanks a lot. Those are interesting for sure.

jon

On Mon, Apr 23, 2007 at 04:53:21PM +0200, Ulf Wiger (TN/EAB) wrote:
> 
> The "Cheap Concurrency" benchmark at 
> http://shootout.alioth.debian.org/debian/benchmark.php?test=message&lang
> =all
> 
> is pretty close to what Joe had, except it stops at
> 3000 threads, IIRC because going higher would break
> too many language implementations (it used to be 500
> threads, but I guess things have improved. See e.g.
> http://erlang.org/ml-archive/erlang-questions/200512/msg00060.html)
> 
> You can play with the shootout page, and make 
> comparisons between different languages. The 
> following view compares Erlang with JDK and 
> Scala (Java-based) and Python (since there
> was another slot).
> 
> http://shootout.alioth.debian.org/debian/fulldata.php?test=message&p1=er
> lang-2&p2=java-2&p3=scala-0&p4=python-0
> 
> 
> BR,
> Ulf W
> 
> > -----Original Message-----
> > From:  
> > [mailto:] On Behalf Of 
> > Jonathan Gold
> > Sent: den 23 april 2007 15:58
> > To: 
> > Subject: [erlang-questions] Any more recent version of Joe's 
> > "PerformanceMeasurements of Threads in Java and Processes in Erlang"
> > 
> > I'm building out some foundational stuff for a project, and 
> > am setting out to explore java speed v. erlang under various 
> > conditions and types of workload.
> > 
> > One of the base conditions is simple processing of messages, 
> > as Joe seems to have investigated in '98 in an Ericsson 
> > report entitled "Performance Measurements of Threads in Java 
> > and Processes in Erlang" located here:
> > 
> >     http://www.sics.se/~joe/ericsson/du98024.html
> > 
> > I'm wondering if there are any more modern renditions of this 
> > experiment? That version uses Erlang 4 and JDK 1.1. I looked 
> > around a bit more and couldn't find very much out there. 
> > 
> > I'm thinking it might be interesting to encapsulate a test 
> > like this (maybe not this exact one, but something similar at 
> > least) in some of the experiments I'm about to do, but want 
> > to see if there are other such reports out there that I 
> > should look at first.
> > 
> > jon
> > _______________________________________________
> > erlang-questions mailing list
> > 
> > http://www.erlang.org/mailman/listinfo/erlang-questions
> > 



More information about the erlang-questions mailing list