SAA-C03 Design High-Performing Architectures Practice Question

A media platform runs a CPU-heavy thumbnail generation workload on an EC2 Auto Scaling group using t3.large instances. During peak traffic, p95 processing time increases significantly even though average CPU remains around 40–50%. CloudWatch also shows CPU credit depletion behavior. Which change will most directly improve performance predictability for this workload?

⚠ Common exam trap

A common mix-up: candidates assume 'CPU credit depletion' can be fixed by increasing credits or scaling out, but the real issue is that burstable instances are fundamentally unsuitable for sustained CPU-heavy workloads, and only switching to a non-burstable instance type (e.g., compute-optimized) guarantees predictable performance.

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

✓

Change the Auto Scaling group instance type to a compute-optimized family (for example, c7i) to provide steady CPU performance.

The t3.large instances rely on CPU credits for burst performance, and when credits are exhausted, CPU performance is throttled to the baseline (e.g., 30% for t3.large). This causes unpredictable processing times during peak traffic, even if average CPU is moderate. Switching to a compute-optimized family like c7i provides dedicated, consistent CPU performance without credit-based throttling, directly improving predictability for CPU-heavy thumbnail generation.

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

✗
Increase the t3.large maximum CPU credits and keep the Auto Scaling group using the same burstable instance type.
Why it's wrong here
Increasing credits can reduce throttling for short or intermittent bursts, but t3 instances still depend on accumulated CPU credits. During sustained peak traffic, credits will deplete and the instance can throttle, so p95 latency predictability remains limited.
✓
Change the Auto Scaling group instance type to a compute-optimized family (for example, c7i) to provide steady CPU performance.
Why this is correct
Compute-optimized instances are designed for consistently high CPU performance and do not rely on a burst-credit model. Switching to a steady-performance family removes the credit-depletion/throttling pattern that is driving the p95 latency spikes under sustained load.
✗
Add a placement group to the existing t3.large instances so they are packed close together for lower latency between nodes.
Why it's wrong here
Placement groups can help certain network-oriented communication patterns by improving latency and/or bandwidth, but the observed symptom is CPU credit depletion on burstable instances. Changing placement does not address CPU throttling caused by credit exhaustion.
✗
Switch the workload to run on Lambda with the same logic so invocations automatically scale without instance selection changes.
Why it's wrong here
Lambda may scale automatically, but provisioned and on-demand execution characteristics do not inherently remove the underlying issue you observed (burstable CPU throttling on t3). For CPU-bound, latency-sensitive batch-like processing, moving to Lambda could introduce other constraints (cold starts, execution limits, concurrency dynamics) and does not directly target credit-based throttling as the root cause.

Go deeper

SAA-C03 Design High-Performing Architectures Practice Question

Why each option matters

Option-by-option breakdown

Related to this question