A web application's performance has been degrading. Historically, the application has had highly- variable workloads, but lately, there has been a steady growth in traffic as the result of a new…

Question

Options

• ABegin using the T2 instance type
• BPurchase more CPU credits for the existing instance
• CIncrease the size of the current instance type
• DConfigure a CloudWatch alarm on the CPU credits metric

• A
  71% (32)
• B
  20% (9)
• D
  9% (4)

Explanation

CPU credits running to zero is a classic symptom of a burstable T-series instance (like T2/T3) being overwhelmed by sustained traffic - these instances earn credits at a baseline rate and burn them during spikes, but steady growth can drain the balance permanently. Option C (increasing instance size) is correct because a larger instance type provides a higher CPU baseline and earns credits faster, directly addressing the sustained load. Option A (T2 instance type with Unlimited mode) allows the instance to burst beyond its credit balance indefinitely by charging for excess CPU usage, preventing throttling during sustained high demand.

B is wrong because CPU credits cannot be purchased - they are earned automatically based on instance size and time. D is wrong because a CloudWatch alarm only notifies you when credits are low; it does not improve performance or prevent throttling.

Memory tip: Think of CPU credits like a bank account - T-series instances earn interest (credits) slowly but can overdraft with Unlimited mode enabled. If your balance keeps hitting zero, either earn more (bigger instance = option C) or remove the overdraft limit (Unlimited = option A).

Note: Option A's phrasing ("begin using the T2 instance type") is ambiguous - on the exam, this likely implies enabling T2/T3 Unlimited, which is the key feature that distinguishes it as a valid fix rather than merely switching to a different T2 size.

No community discussion yet for this question.