A proof-of-concept tuning exercise is being written up as a governed optimisation plan. What should be reviewed before any change to production behaviour is made?
Select an answer to reveal the explanation.
Short Explanation
You review where the time actually goes, how data is touched, and what logic repeats. That's what you optimise, in that order.
Full Explanation
The correct review target is where the workload's hotspots are, how it accesses data, and where logic is repeated unnecessarily. Optimisation that is not grounded in where time is actually spent, how records are accessed, and where logic duplicates itself tends to change code that was never the bottleneck. Reviewing the wording of the code's inline comments is incorrect because comments do not affect runtime behaviour and reviewing them does not identify a hotspot. Reviewing the number of files in the affected module is incorrect because file count says nothing about where execution time or repeated work actually occurs. Reviewing the initials of the original developer who wrote the code is incorrect because authorship has no bearing on where the workload spends its time. For the exam, name hotspots, data access patterns, and repeated logic as the three review targets.