Why should data models all be accelerated? What about the built-in Data Models?
I disagree with the premise that all data models should be accelerated. One should only accelerate the data models one uses.
Pros of acceleration: faster access to data
Cons: additional storage; additional scheduled searches