This illustration shows the decision tree for choosing a data source type. At the top, there is a box with the question "Require JTA (for example, EJB)"? If the answer is no, then a native data source should be used. If the answer is yes, then proceed to the next box, which contains the question "Require two-phase commit"? If yes, then a nonemulated data source should be used; if yes, then an emulated data source should be used.