I'm asked over and over again by my customers whether glassfish is stable and scalable enough for real world projects. Although in my experience it is the case (there are of course some bugs, but I do not know a bug-free appserver :-)), I searched for a more reliable source to answer this question. What I missed was the post "Bragging Time: 883.66 JOPS@Standard on GlassFish V2". This post was also well commented by Scott Oak. It seems like glassfish outperformed BEA these days, however the difference was rather small. Such small performance differences actually doesn't matters in an average real world project, however it is a good reference to the customers.
Also interesting are the results. Even the settings for the JVM were publishes - which is interesting for the practice :-).
airhacks.fm the podcast:
Stay in touch: airhacks.news.