SAA-C03 Design Cost-Optimized Architectures Practice Question
A team runs an EC2-based API on a single Auto Scaling group (ASG). Over the last month, they observed: - Average CPU utilization is ~15%. - p95 latency is stable and within the performance target. - The attached EBS volumes are gp3, provisioned with high baseline IOPS/throughput “just to be safe,” but CloudWatch shows consistently low utilization of those provisioned IOPS/throughput limits. They want to reduce monthly cost while maintaining current performance. Which action is the best cost-optimized choice?
⚠ Common exam trap
Watch out — candidates often assume EBS performance settings are fixed or risky to change, or that scaling out the ASG is always the best cost optimization, when in fact gp3 allows flexible, no-downtime IOPS/throughput adjustments and the real savings come from matching provisioned resources to actual utilization.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Right-size both the compute and the gp3 volumes: reduce the EC2 instance size (via the ASG launch template/desired capacity configuration) and update gp3 IOPS/throughput settings to match observed utilization while keeping p95 latency targets.
The workload is over-provisioned in both compute and storage. Average CPU is only 15%, so a smaller instance size can handle the load without affecting p95 latency. The gp3 volumes have high baseline IOPS/throughput that are never used, so reducing them to match actual utilization directly lowers costs without performance risk.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Stop resizing EBS and only scale out the ASG during peak traffic, because changing EBS performance settings risks latency spikes.
Why it's wrong here
Scaling out only during peak traffic would add EC2 instances and therefore increase compute spend, and any new instances launched from the same launch template would inherit the same overprovisioned gp3 IOPS/throughput, multiplying the existing waste rather than removing it. It also leaves the currently deployed volumes paying for performance they consistently do not use. Because gp3 IOPS and throughput can be modified live without resizing or downtime, and because p95 latency is already stable, gradually reducing these settings to match observed utilization does not introduce a meaningful latency-spike risk.
- ✓
Right-size both the compute and the gp3 volumes: reduce the EC2 instance size (via the ASG launch template/desired capacity configuration) and update gp3 IOPS/throughput settings to match observed utilization while keeping p95 latency targets.
Why this is correct
The metrics indicate headroom that is not being used (low CPU, stable latency, and low gp3 utilization). The most direct cost optimization is to reduce overprovisioned spend by right-sizing the instance type and tuning gp3 IOPS/throughput to match actual demand. Because performance and latency are already stable, these changes are the most likely to reduce cost without degrading performance.
- ✗
Switch the instances to EC2 Spot immediately, because Spot always lowers costs without adding operational risk or affecting performance.
Why it's wrong here
Spot can reduce costs, but it introduces possible interruptions and capacity variability. The prompt does not state the workload can safely handle interruptions, and it does not guarantee that performance will remain stable under Spot interruptions/replacements.
- ✗
Move the workload to a larger instance class and keep the gp3 settings unchanged to avoid operational tuning work.
Why it's wrong here
Moving to a larger instance class directly raises the per-hour compute cost and provides additional CPU capacity that the low utilization and stable p95 latency do not demand. Keeping the gp3 settings unchanged preserves the recurring expense of unused provisioned IOPS and throughput, so you remain overprovisioned on both compute and storage. The effort of modifying gp3 volume settings is negligible because changes are applied dynamically and non-disruptively, so avoiding that small tuning task does not justify accepting higher infrastructure bills.
Go deeper
Related to this question
About these practice questions
This SAA-C03 question is part of Courseiva's 935-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This SAA-C03 practice question is part of Courseiva's free Amazon Web Services certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the SAA-C03 exam.