Okay/understood, thanks Leonardo. :) >> The most interesting fact is the decrease in the throughput for Spring MVC.
I found that interesting too. >> If you think that using Spring MVC-JSP will give you a better performance, think twice, because MyFaces with a CDI implementation like OpenWebbeans will do a better job. And I was glad to see this mentioned... I'm definitely using all of the above (MyFaces Core + OpenWebBeans + CDI via TomEE 1.6.0 snapshot). :) I think I have been using MyFaces Core ever since August or September 2012, and I have been very satisfied with the performance, and GC does not seem to be an issue in/with my app as my app has been running perfectly at -Xms1250m -Xmx1250m -server. Honestly, I don't ever remember experiencing an OutOfMemory issue with my app. :) >> Servlet-JSP uses the least possible amount of memory, but the surprising part is MyFaces uses 32% less memory than Spring MVC-JSP, and in the best case the view pool uses 46% less memory. That means in some cases under high load and low memory, MyFaces will probably perform better than Spring MVC-JSP due to the garbage collector effect. Very interesting. Prior to reading this blog of yours, I was not aware of how well Servlet-JSP performs compared to JSF, and now I'm glad to see Spring MVC JSP included in the comparisons as well. I will be honest with you, when I look at Instances in jvisualvm, I see more MyFaces Core instances than instances of my classes, and the MyFaces Core instances seem to take up more memory than instances of my classes. >> This feature is ideal for viewing the session size—just look for org.apache.catalina.session.StandardSession retained size While reading, I was just about to ask about this, and now I know. Since I'm not using Yourkit, I hope I can see the session size in jvisualvm (or Netbeans profiler). Anyway, very nice-n-informative blog/post/article. Thank you! On Wed, Jul 17, 2013 at 10:13 AM, Leonardo Uribe <lu4...@gmail.com> wrote: > Hi Howard > > JSF 2.1 is java 1.6 compatible. In this case, an improvement in the JVM > will affect all frameworks in the same way, so a change using java 7 or 8 > will not affect the relative differences. > > Regards, > > Leonardo. > On Jul 17, 2013 1:32 PM, "Howard W. Smith, Jr." <smithh032...@gmail.com> > wrote: > > > First of all, kudos for the following blog/post: > > > > JSFCentral - Understanding JSF 2.0 Performance - Part 3[1] > > > > I started reading this, and not done yet. > > > > >> Instead, it's more interesting to check the ability of a web framework > > to deal with effects like concurrency and get an idea of the overhead > > involved in using a web framework against the fastest possible solution > in > > Java. > > > > So far, I'm seeing, possibly, latest releases of open source software > used > > in the test...except for the JDK. Does 'fastest possible solution in > Java' > > = JDK6.0.30? > > > > JDK7 has been available for quite some time and I know I'm using JDK7 > > (latest version, always), and JDK8 is out there and around the corner. > > > > Is there any reason why Java = JDK6.0.30 in the tests? > > > > > > [1] > > http://www.jsfcentral.com/articles/understanding_jsf_performance_3.html > > >
