Question #1
A SaaS company hosts its CRM application on an Auto Scaling group that uses a mix of On-Demand EC2 instances of varying types and sizes. The application experiences heavy usage during regular business hours (9 AM to 5 PM). However, users have reported noticeable slowness during the initial hours of the workday, after which performance returns to normal.
What is the MOST operationally efficient way to ensure the application performs well right from the beginning of the day?
A. Set up a Predictive scaling policy for the Auto Scaling group to proactively manage instance counts.
B. Set up a Scheduled scaling policy to pre-launch instances before peak hours begin.
C. Use a Dynamic scaling policy based on memory usage to scale the Auto Scaling group.
D. Use a Dynamic scaling policy based on CPU usage to scale the Auto Scaling group.
Show answer
✅ Correct Answer: B. Set up a Scheduled scaling policy to pre-launch instances before peak hours begin.
🧾 Explanation:
The issue occurs at a predictable time every day—right at the start of office hours—so a Scheduled scaling policy is the most efficient approach. It allows you to scale out before the traffic spike begins, ensuring the application is already ready to handle demand.
- Predictive scaling (A) could help, but it adds complexity and is best suited for workloads with highly variable and non-obvious patterns.
- Dynamic scaling based on memory (C) isn’t directly supported by default metrics in Auto Scaling.
- Dynamic scaling based on CPU (D) only reacts after a resource bottleneck is detected, which doesn’t prevent initial performance issues.
Scheduled scaling offers simplicity and precision for known traffic patterns like this one.
Question #2
You’re managing an application that runs on a fleet of EC2 instances in an Auto Scaling group. To enhance its responsiveness to load changes, you need to implement a scaling strategy that adjusts the number of instances based on specific metric thresholds. You also want the scaling process to respond with different levels of adjustment depending on how far the metric deviates from the set threshold.
Which type of Auto Scaling policy is BEST suited for this requirement?
A. Scheduled scaling
B. Simple scaling
C. Step scaling
D. Target tracking scaling
Show answer
✅ Correct Answer: C. Step scaling
🧾 Explanation:
Step scaling is designed for scenarios where you want scaling actions to vary depending on how much a metric deviates from a target value. For example, you might add 1 instance if CPU is 60%, but add 3 instances if it reaches 80%. This offers granular control and is triggered by CloudWatch alarms.
- Scheduled scaling (A) is used when you need to scale at specific times, not based on metrics.
- Simple scaling (B) is a more basic version that triggers a single scaling action per alarm.
- Target tracking scaling (D) keeps a metric like CPU utilization at a target value, but doesn’t offer the same level of custom step adjustments.
Step scaling gives you both flexibility and control over how your app scales based on the severity of the metric.
Question #3
Your engineering team has built a new Amazon Machine Image (AMI) for your web application. The application currently runs within an Auto Scaling group using a defined launch template. You need to ensure that any newly launched instances use this updated AMI going forward.
What is the MOST appropriate step to take?
A. Create a new target group.
B. Create a new target group and a new launch template.
C. No changes are necessary. The Auto Scaling group will automatically use the latest AMI.
D. Create a new launch template or a new version of the existing template.
Show answer
✅ Correct Answer: D. Create a new launch template or a new version of the existing template.
🧾 Explanation:
Amazon EC2 Auto Scaling groups use a launch template or launch configuration to determine how to provision new instances. This includes settings like instance type, key pair, and crucially—the AMI ID.
- If you want to start using a new AMI, you need to either:
- Create a new launch template, or
- Create a new version of the existing template with the updated AMI.
Then, update your Auto Scaling group to use the new version.
C (Do nothing) is incorrect — the Auto Scaling group doesn’t automatically pick up new AMIs unless the launch template or configuration is updated.
A (Create a new target group) is incorrect — target groups are used for routing traffic with a load balancer, not for defining instance configurations.
B (Both target group and launch template) is unnecessary — target groups aren’t involved in launching instances.
Question #4
A web application is hosted in an Auto Scaling group of EC2 instances. Every morning, it experiences a surge in traffic, and users report request timeouts. It takes about 1 minute for a new EC2 instance to become fully operational after launch.
How should a Solutions Architect redesign the architecture to better handle this traffic pattern?
A. Deploy a Network Load Balancer with slow-start mode
B. Create a new launch template with a larger instance type
C. Implement a step scaling policy and configure the instance warm-up time
D. Set up a CloudFront distribution with the EC2 instance as the origin
Show answer
✅ Correct Answer: C. Implement a step scaling policy and configure the instance warm-up time
Explanation:
Auto Scaling may route traffic to new EC2 instances before they’re ready. Setting an instance warm-up time ensures instances are only considered available after they’re fully initialized, reducing timeouts during traffic spikes. Step scaling adds further control over scaling behavior.
Question #5
A commercial bank runs a forex trading application using an Auto Scaling group to maintain performance and cost-efficiency. They want to prevent rapid scaling actions that might cause instability or system unavailability.
Which of the following statements about Auto Scaling cooldown periods are TRUE? (Select TWO.)
A. Its default value is 300 seconds
B. It ensures that the Auto Scaling group does not launch or terminate additional EC2 instances before the previous scaling activity takes effect
C. It ensures that the Auto Scaling group launches or terminates additional EC2 instances without any downtime
D. It ensures that before the Auto Scaling group scales out, the EC2 instances have an ample time to cooldown
E. Its default value is 600 seconds
Show answer
✅ Correct Answers:
A. Its default value is 300 seconds
B. It ensures that the Auto Scaling group does not launch or terminate additional EC2 instances before the previous scaling activity takes effect
🧾 Explanation:
Option E is incorrect because 600 seconds is not the default value.
The cooldown period is a mechanism that helps prevent over-scaling or rapid fluctuations by pausing additional scaling actions until the effects of a previous one are in place.
The default cooldown is 300 seconds (5 minutes).
It does not guarantee zero downtime (C ❌) and it’s not specific to scaling out only (D ❌).