On Saturday 28 August 2010 00:11:11 Rainer Jung wrote: > On 27.08.2010 21:58, Wesley Acheson wrote: > > On Fri, Aug 27, 2010 at 9:41 PM, Pid<p...@pidster.com> wrote: > >> On 27/08/2010 18:51, Wesley Acheson wrote: > >>> I think the reason for doing this in ruby is that ruby is single > >>> threaded, I've been told. The JVM isn't.
> Adding unqualified rumors: Ruby is not single-threaded, Right - the language has a thread concept that can be explored as true multi- threading by interpreters. But the original C-based ruby interpreters for do have a global interpreter lock that causes only one native thread to be running within one ruby process. This makes these interpreters effectively single threaded. (Ruby up to 1.8 only used 1 native thread at all, Ruby 1.9 maps ruby threads to multiple native threads, but still has the global interpreter lock.) > but the Rails > framework has a huge lock that effectively make the biggest part of > request handling serialized. Usually Ruby webapps are based on Rails. So > yes, Ruby on Rails needs multiple server processes in parallel to > effectively scale. That might be an outdated rumor though. It is in so far outdated that Rails from 2.3 on has a threadsafe configuration option that enables multi-threaded request processing. This needs a multi- threaded runtime though to make any sense. JRuby is (of course) multi- threaded, I'm not sure if there is any other Ruby interpreter that is multi- threaded. > Regards, > > Rainer Rainer --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
