A governed modernisation plan growing out of a proof of concept must choose between three architectural extremes. Which approach is defensible?
Select an answer to reveal the explanation.
Short Explanation
Single pattern applied everywhere is a shortcut, not a decision. You need to evaluate each module against its business and technical constraints, then use a modular service, an API, or a hybrid-cloud pattern only where it fits.
Full Explanation
The correct approach is to evaluate each module against its business and technical constraints and apply a modular service, an API, or a hybrid-cloud pattern only where it fits. The governed plan evaluates modules individually against real constraints rather than committing to one pattern across the board, because different modules have different coupling, scaling and ownership needs.
Applying the same target pattern—microservices, APIs, or hybrid-cloud—to every module regardless of its individual constraints is incorrect because uniform application ignores the differences between modules that made per-module evaluation necessary in the first place.
Choosing the architecture that requires the least documentation to produce is incorrect because documentation effort is not a valid basis for an architectural decision.
Choosing the architecture the proof of concept happened to use, without reassessing it for production scale, is incorrect because the proof of concept is deliberately narrow in scope and its choices are not automatically valid at production scale.
For the exam, expect 'evaluate per module against constraints' to beat any single blanket pattern.