C100DBA Indexing and Performance 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.

• ✗

  Hashed indexes provide no benefit for equality queries on high-cardinality fields and should not be used.

  Why it's wrong here

  This overstates the case. Hashed indexes distribute values evenly, which is valuable for sharding and for equality queries where the goal is even data distribution. However, for a single-node equality lookup on a high-cardinality field, a B-tree index is generally more efficient because it supports range queries and ordered scans. Saying hashed indexes provide no benefit at all is too absolute and ignores their role in sharding.

• ✗

  A hashed index will improve equality query performance but will prevent the use of range queries on status.

  Why it's wrong here

  While it is true that hashed indexes do not support range queries, the scenario is about equality lookups on a high-cardinality field. The statement implies a performance improvement, which is not the primary benefit here. A B-tree index would already handle equality efficiently and also support ranges. The DBA should not expect a hashed index to improve equality query performance; it may even be less efficient due to hashing overhead.

• ✓

  A hashed index will not improve equality query performance and is primarily intended for even data distribution in sharded clusters.

  Why this is correct

  Hashed indexes are designed to distribute documents evenly across shards by hashing the shard key value. For a single-node equality lookup, a standard B-tree index is usually sufficient and more versatile because it supports range queries and sorted access. Since the field already has high cardinality, a B-tree index on status would provide efficient equality lookups without the drawbacks of hashed indexes, which cannot be used for range queries.

• ✗

  Hashed indexes are ideal for equality lookups and will outperform a standard B-tree index for this query pattern.

  Why it's wrong here

  Hashed indexes are designed to distribute values uniformly, not to maximize single-key lookup performance. For a simple equality query on a high-cardinality field, a standard B-tree index typically performs as well or better because it stores keys in sorted order and can serve range queries too. The claim that hashed indexes will outperform B-tree indexes for this pattern is not supported; hashed indexes mainly benefit sharding and even distribution.