The original batch subsystem will not work well with capabilities and requirements. It's also fairly limited in the validation it can offer for the job repositories.
The job-repository-type will be removed as well as the job-repository=jdbc. A new jdbc-job-repository and in-memory-job-repository resource will be added. There will also be a default-job-repository which will accept a name from one of the defined job repositories.
The jndi-name will no longer be used and a data-source name will be used for the JDBC job repository. For example resource value might be ExampleDS.
A new deployment descriptor will be added to allow a user to defined a job repository configured on the subsystem.
Due to these changes the persisted XML will likely need some small alterations. Mainly ensuring a name attribute is persisted and multiple repositories are allowed to be defined.
The current batch subsystem will remain. It will allow the server to boot with old configurations, but the default with be the new batch-jberet subsystem.