Deferred testing is testing you put off until other testing has been completed. This would possibly be due to its being a lower priority, or its functionality is not completely coded yet, or maybe even because it is a very long or complicated test which requires a dedicated amount of time without any places to pause.
Success is the ability to go from one failure to another with no loss of enthusiasm.
~ Winston Churchill ~
Um, not necessarily, Rich. It depends (I can no longer use this word without wincing) on the context of "deferred".
I've had to test an application known as "deferred". The application wrote certain transactions to a "deferred" (temporary) database, and updated the system on a scheduled basis.
It also depends on the context for cluster testing. If you're testing a clustered configuration, for example, you're ensuring messages get routed correctly based on source and load, and responses are returned correctly based on the same. You'll also want to test failover procedures.
These are just two definitions, however, it really depends on what your context is, koti. Do you have more information?