A company has deployed a CRM application on IBM Cloud, which records information about customer sales, as well as customer interaction and the channels they use. The marketing department wants to…

Question

Options

• AImplement a read-replica of the PostgreSQL database and make it the target for the reporting
• BExport the relevant PostgreSQL tables into YAML files and use Data Engine to query them
• CAdd a new member to the PostgreSQL instance cluster and mark it as reportmaster
• DEnsure that the PostgreSQL instance is in the closest MZR to the marketing department

• A
  74% (26)
• B
  3% (1)
• C
  9% (3)
• D
  14% (5)

Explanation

Read replicas solve the core problem here: they create a synchronized copy of the primary database that absorbs read-heavy workloads (like complex reporting queries) without competing for resources on the primary instance that serves live application users. Option A is correct because directing reporting jobs to a read-replica isolates that I/O-intensive work, preserving response times for CRM users hitting the primary database.

Why the distractors fail:

• B is wrong because IBM Cloud Databases for PostgreSQL data isn't exported as YAML, and IBM Data Engine (Presto-based) queries object storage, not a live relational export workflow like this.
• C is invented - PostgreSQL has no concept of a "reportmaster" role; you can add read replicas, but the term and process described don't exist in IBM Cloud or standard PostgreSQL.
• D is a network latency optimization, not a database load-isolation strategy - moving the instance closer to the marketing team does nothing to reduce the performance impact on other users sharing the same database.

Memory tip: Think "reads go to replicas, writes go to primary." Whenever an exam scenario describes heavy read workloads (reporting, analytics, dashboards) threatening write/OLTP performance, a read-replica is almost always the correct mitigation.

No community discussion yet for this question.