A design document for a new watsonx.data deployment lists buckets, catalogs and engines, together with how they will be associated, but says nothing about who may access which tables. What is missing?
Select an answer to reveal the explanation.
Short Explanation
The document only maps what connects to what. You still need the governance layer — access policies and the roles that will hold them — because association gives reachability, not permission.
Full Explanation
The missing piece is the governance layer — access policies and the roles that will hold them. Governance is a distinct component of the platform. The complete design has to state which roles exist, which principals hold them, and what policies apply to catalogs, schemas and tables, because that is what determines the rows any particular user sees. The claim that nothing is missing because catalog association is itself the access control mechanism is wrong and is a common and consequential confusion. Association is topology: it makes a catalog reachable from an engine. Once reachable, what any individual may read is decided entirely by access policy, so a design with associations and no policies has connected everything to everyone it can. The engine sizing detail is wrong because sizing governs throughput and concurrency, not entitlement. A larger engine lets more people query; it does not decide what they are allowed to query. The object storage lifecycle rules are wrong because lifecycle rules manage how long objects persist and on what tier. Retention is a related governance concern but it is not access control, and a deleted-after-90-days rule says nothing about who may read the data today. For the exam, keep topology and entitlement separate: association is reachability, policy is permission.