Question #1614
A company hosts its applications on Amazon EC2 instances within Auto Scaling groups. The applications face abrupt traffic spikes that occur unpredictably at irregular intervals throughout the month. The company needs to ensure consistent application performance during these sudden traffic surges while optimizing costs.
Which solution MOST cost-effectively fulfills these requirements?
Manually adjust the Auto Scaling group size based on observed traffic patterns.
Implement forecast-based scaling to anticipate traffic changes and adjust the Auto Scaling group size.
Configure real-time metric-driven scaling to dynamically adjust the Auto Scaling group size.
Define time-bound scaling policies to adjust the Auto Scaling group size at specific intervals.
Explanation
Option C is correct because real-time metric-driven scaling (e.g., using Amazon CloudWatch alarms for CPU or network usage) allows the Auto Scaling group to respond immediately to sudden traffic spikes. This ensures consistent performance by scaling out when demand increases and scaling in when demand drops, optimizing costs by only provisioning resources when needed.
Option A is incorrect because manual scaling cannot react quickly enough to abrupt, unpredictable traffic changes. Option B (forecast-based scaling) relies on predictable patterns, which the question explicitly states do not exist. Option D (time-bound scaling) assumes traffic follows a fixed schedule, which is incompatible with the irregular intervals described.
Key Points:
- Real-time metrics (e.g., CPU, network) enable dynamic scaling for unpredictable workloads.
- Cost optimization is achieved by scaling only when necessary.
- Manual, forecast-based, or scheduled scaling are unsuitable for irregular traffic spikes.
Answer
The correct answer is: C