C100DEV Data Modeling Practice Question

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  Apply the Bucket Pattern by grouping follower ObjectIds into fixed-size buckets of 1000 per document in a separate collection.

  Why it's wrong here

  The Bucket Pattern is intended for time-series or high-volume event data where grouping measurements by time window reduces document count and index size. Followers are not time-bucketed measurements; applying this pattern would force awkward pagination logic and would not solve the underlying problem of a single array exceeding 16MB in the outlier documents, since the original user document would still hold the embedded list.

• ✓

  Apply the Outlier Pattern by adding an overflow flag and moving the followers array into a separate collection only for users that exceed a threshold.

  Why this is correct

  The Outlier Pattern handles exactly this case: most documents fit comfortably in the common shape, but a few outlier documents would exceed limits. By flagging outliers (for example with a 'has_extra' boolean) and storing the overflow in a companion collection, the schema stays efficient for the majority while celebrities are handled correctly without redesigning the whole model.

• ✗

  Apply the Extended Reference Pattern by duplicating the ten most important follower documents into the user document.

  Why it's wrong here

  The Extended Reference Pattern copies a small set of frequently needed fields from a referenced document to avoid extra lookups, which is useful for product or order summaries. Duplicating follower profile data into each user document does not resolve the 16MB overflow for celebrities and would introduce stale duplicated data that must be kept in sync, adding write overhead without fixing the size problem.

• ✗

  Apply the Subset Pattern by storing only the 100 most recent followers in the user document and the full list in a separate collection.

  Why it's wrong here

  The Subset Pattern is designed for cases where a small, frequently accessed subset of a large array (such as the 10 most recent reviews) can satisfy the majority of reads while the full set lives elsewhere. For followers, applications typically need to paginate through the entire list or check membership for any given user, so truncating to a recent subset would break follower lookups and give misleading counts.