AWS Certified Database – Specialty — Question 345

A pharmaceutical company's drug search API is using an Amazon Neptune DB cluster. A bulk uploader process automatically updates the information in the database a few times each week. A few weeks ago during a bulk upload, a database specialist noticed that the database started to respond frequently with a
ThrottlingException error. The problem also occurred with subsequent uploads.
The database specialist must create a solution to prevent ThrottlingException errors for the database. The solution must minimize the downtime of the cluster.
Which solution meets these requirements?

Answer options

• A. Create a read replica that uses a larger instance size than the primary DB instance. Fail over the primary DB instance to the read replica.
• B. Add a read replica to each Availability Zone. Use an instance for the read replica that is the same size as the primary DB instance. Keep the traffic between the API and the database within the Availability Zone.
• C. Create a read replica that uses a larger instance size than the primary DB instance. Offload the reads from the primary DB instance.
• D. Take the latest backup, and restore it in a DB cluster of a larger size. Point the application to the newly created DB cluster.

Correct answer: A

Explanation

Creating a larger read replica and initiating a failover is the most efficient way to scale up an Amazon Neptune primary instance with minimal downtime, as the failover process typically completes in less than 30 seconds. Offloading reads to a larger replica (Option C) does not address the write-heavy ThrottlingException errors caused by bulk uploads on the primary instance. Restoring from a backup (Option D) would take significantly longer and result in much higher downtime than a replica failover.