Performance impact of injecting javax.ejb.Singleton vs javax.inject.Singleton

[troup]

I am building a Java EE 6 web application running on JBoss AS 7.1.1 final. In the app we have defined a Singleton EJB and injected that in to another class. (I know that javax.ejb.Singleton is not part of the CDI spec but please bear with me.)

We noticed a performance issue in our code whereby method invokations on the injected object took 60ms to complete where we saw them as being near instantaneous when run against a non-injected instance.

When we changed the injected class to be  a javax.inject.Singleton we saw a massive speed up in our code. Back to near instantaneous execution.

Can anyone tell me why method execution on an injected javax.ejb.Singleton is so badly affected?

[atomicknight]

By default, singleton session beans are transactional (section 13.3.7 of the EJB 3.1 specification) and require acquisition of an exclusive lock for every business method invocation (sections 4.8.5.4 and 4.8.5.5). In contrast, a javax.inject.Singleton is not transactional and does not support container-managed concurrency (the major consequence being that no locking scheme is implemented by the container).

If you annotate your singleton session bean with @TransactionAttribute(NOT_SUPPORTED) and @Lock(READ), you should see significantly better performance, though there may still be some overhead. If you don't need EJB features, stick with @ApplicationScoped (javax.inject.Singleton is not defined by CDI, and its semantics are therefore not governed by that specification).

[troup]

Thanks for the info. I had already annotated the singleton session bean with @Lock(READ) but not @TransactionAttribute(NOT_SUPPORTED). Since I don't need EJB features I will stick to @ApplicationScoped.