SAA-C03 Practice Question 59: Aurora IO Optimized Cost
AWS Certified Solutions Architect – Associate (SAA-C03) · Test 6 · question 9 of 10
An Aurora PostgreSQL cluster serves a write-heavy analytics ingest workload. Cost Explorer shows that charges for I/O operations are about 40% of everything the company spends on the cluster, and the monthly figure swings unpredictably with ingest volume. The team wants a lower and more predictable bill without changing the application or accepting worse performance.
What should the solutions architect do?
Show answer and explanation
Answer:
Aurora offers two cluster storage configurations. Aurora Standard bills instances and storage plus a rate per million I/O requests, which is where the volatile 40% is coming from. Aurora I/O-Optimized removes the per-request I/O charge entirely, you pay only for instance usage and storage, in exchange for a higher instance and storage rate. AWS states that I/O-Optimized is the better choice once I/O spending reaches 25% or more of total Aurora spending, and at 40% this cluster is well past that line. Switching is a cluster setting, so no application change is required, and removing the per-request component is what makes the bill predictable.
Option A changes how compute is provisioned, not how I/O is billed. Aurora Serverless v2 clusters are still subject to the same storage configuration choice, and a low minimum ACU on a sustained write-heavy workload would throttle performance, which the requirement forbids.
Option C attacks the wrong side of the workload. A cache in front of the database reduces *read* traffic, the ingest here is write-heavy, and introducing a cache is exactly the application change that was ruled out.
Option D reduces the instance line item, which is real but separate. Reserved Instances do not discount I/O requests at all, so the 40% of spend identified in the question is completely untouched and stays just as unpredictable.
Reference: Amazon Aurora storage