Amazon AWS Certified Solutions Architect - Associate SAA-C03 Bundle

The option Review minimum billable object size and minimum storage duration addresses a separate condition: millions of 5 KB objects are proposed for Glacier transition after one day. Tiny files are often poor cold-tier candidates. The actual requirement is files older than seven years must expire unless held for litigation. Recommended corrective action: Apply hold-aware lifecycle expiration and audit enforcement. Storage policy must consider retention exceptions.

Archive eligible masters in a Glacier retrieval tier with planned restore time would help if video source files are revisited once each year for re-editing. Cold archive fits rare restoration. That is not the symptom described here: files older than seven years must expire unless held for litigation. The matching response is Apply hold-aware lifecycle expiration and audit enforcement, because storage policy must consider retention exceptions.

Use a frequently accessed S3 tier and CDN caching solves a different operational problem where a catalogue thumbnail is fetched millions of times per day. Hot objects should avoid retrieval charges and origin re-fetches. It does not directly address files older than seven years must expire unless held for litigation. The case calls for Apply hold-aware lifecycle expiration and audit enforcement, since storage policy must consider retention exceptions.

Schedule nonproduction instance shutdown and startup would help if a dev environment runs every night despite no users. Stopping idle EC2 compute reduces usage charges. That is not the symptom described here: a steady 24/7 EC2 baseline remains in On-Demand with high predictable usage. The matching response is Purchase suitable Compute Savings Plans or EC2 Instance Savings Plans after commitment analysis, because savings Plans discount committed hourly compute usage.

For the alternative failure an unpredictable bursty web service pays for a permanently overprovisioned peak fleet, the control Use Auto Scaling to shrink On-Demand capacity outside peaks is relevant: elastic scaling reduces unused instance-hours. Here the evidence instead is a steady 24/7 EC2 baseline remains in On-Demand with high predictable usage. The selected technical mechanism is Purchase suitable Compute Savings Plans or EC2 Instance Savings Plans after commitment analysis: savings Plans discount committed hourly compute usage.

The competing option Right-size instances using measured CPU/memory/network demand would be useful when an instance family is twice as powerful as the workload needs. Right sizing reduces excess provisioned compute. The present issue is a steady 24/7 EC2 baseline remains in On-Demand with high predictable usage; it is more directly addressed by Purchase suitable Compute Savings Plans or EC2 Instance Savings Plans after commitment analysis. Savings Plans discount committed hourly compute usage.

For the alternative failure an unpredictable bursty web service pays for a permanently overprovisioned peak fleet, the control Use Auto Scaling to shrink On-Demand capacity outside peaks is relevant: elastic scaling reduces unused instance-hours. Here the evidence instead is a noncritical batch worker can be interrupted without losing business correctness. The selected technical mechanism is Use Spot Instances with checkpoints and interruption handling: spot offers discounts in exchange for interruption risk.

The competing option Schedule nonproduction instance shutdown and startup would be useful when a dev environment runs every night despite no users. Stopping idle EC2 compute reduces usage charges. The present issue is a noncritical batch worker can be interrupted without losing business correctness; it is more directly addressed by Use Spot Instances with checkpoints and interruption handling. Spot offers discounts in exchange for interruption risk.

The option Right-size instances using measured CPU/memory/network demand addresses a separate condition: an instance family is twice as powerful as the workload needs. Right sizing reduces excess provisioned compute. The actual requirement is a noncritical batch worker can be interrupted without losing business correctness. Recommended corrective action: Use Spot Instances with checkpoints and interruption handling. Spot offers discounts in exchange for interruption risk.

Use Spot Instances with checkpoints and interruption handling solves a different operational problem where a noncritical batch worker can be interrupted without losing business correctness. Spot offers discounts in exchange for interruption risk. It does not directly address an unpredictable bursty web service pays for a permanently overprovisioned peak fleet. The case calls for Use Auto Scaling to shrink On-Demand capacity outside peaks, since elastic scaling reduces unused instance-hours.

The competing option Schedule nonproduction instance shutdown and startup would be useful when a dev environment runs every night despite no users. Stopping idle EC2 compute reduces usage charges. The present issue is an unpredictable bursty web service pays for a permanently overprovisioned peak fleet; it is more directly addressed by Use Auto Scaling to shrink On-Demand capacity outside peaks. Elastic scaling reduces unused instance-hours.

Right-size instances using measured CPU/memory/network demand would help if an instance family is twice as powerful as the workload needs. Right sizing reduces excess provisioned compute. That is not the symptom described here: an unpredictable bursty web service pays for a permanently overprovisioned peak fleet. The matching response is Use Auto Scaling to shrink On-Demand capacity outside peaks, because elastic scaling reduces unused instance-hours.

The competing option Use Spot Instances with checkpoints and interruption handling would be useful when a noncritical batch worker can be interrupted without losing business correctness. Spot offers discounts in exchange for interruption risk. The present issue is a dev environment runs every night despite no users; it is more directly addressed by Schedule nonproduction instance shutdown and startup. Stopping idle EC2 compute reduces usage charges.

The option Right-size instances using measured CPU/memory/network demand addresses a separate condition: an instance family is twice as powerful as the workload needs. Right sizing reduces excess provisioned compute. The actual requirement is a dev environment runs every night despite no users. Recommended corrective action: Schedule nonproduction instance shutdown and startup. Stopping idle EC2 compute reduces usage charges.

Use Auto Scaling to shrink On-Demand capacity outside peaks would help if an unpredictable bursty web service pays for a permanently overprovisioned peak fleet. Elastic scaling reduces unused instance-hours. That is not the symptom described here: a dev environment runs every night despite no users. The matching response is Schedule nonproduction instance shutdown and startup, because stopping idle EC2 compute reduces usage charges.

The option Use Auto Scaling to shrink On-Demand capacity outside peaks addresses a separate condition: an unpredictable bursty web service pays for a permanently overprovisioned peak fleet. Elastic scaling reduces unused instance-hours. The actual requirement is an instance family is twice as powerful as the workload needs. Recommended corrective action: Right-size instances using measured CPU/memory/network demand. Right sizing reduces excess provisioned compute.

For the alternative failure a noncritical batch worker can be interrupted without losing business correctness, the control Use Spot Instances with checkpoints and interruption handling is relevant: spot offers discounts in exchange for interruption risk. Here the evidence instead is an instance family is twice as powerful as the workload needs. The selected technical mechanism is Right-size instances using measured CPU/memory/network demand: right sizing reduces excess provisioned compute.

Purchase suitable Compute Savings Plans or EC2 Instance Savings Plans after commitment analysis solves a different operational problem where a steady 24/7 EC2 baseline remains in On-Demand with high predictable usage. Savings Plans discount committed hourly compute usage. It does not directly address an instance family is twice as powerful as the workload needs. The case calls for Right-size instances using measured CPU/memory/network demand, since right sizing reduces excess provisioned compute.

Benchmark EC2/ECS continuous compute economics against Lambda would help if a workload has stable continuous high CPU load and Lambda invocation charges exceed EC2. Long steady compute may favor provisioned capacity. That is not the symptom described here: a low-traffic webhook service runs on an always-on EC2 instance. The matching response is Migrate supported short handlers to Lambda with API Gateway, because serverless per-invocation billing can suit sporadic traffic.

For the alternative failure a Lambda function waits idle on a slow synchronous external request, the control Make independent long operations asynchronous where business semantics allow is relevant: asynchrony avoids billed idle request time. Here the evidence instead is a low-traffic webhook service runs on an always-on EC2 instance. The selected technical mechanism is Migrate supported short handlers to Lambda with API Gateway: serverless per-invocation billing can suit sporadic traffic.

The competing option Schedule ECS service desired-count reductions with safeguards would be useful when an ECS service's tasks run 24/7 at full size although demand is only daytime. Scaling down idle containers reduces compute charges. The present issue is a low-traffic webhook service runs on an always-on EC2 instance; it is more directly addressed by Migrate supported short handlers to Lambda with API Gateway. Serverless per-invocation billing can suit sporadic traffic.

For the alternative failure a Lambda function waits idle on a slow synchronous external request, the control Make independent long operations asynchronous where business semantics allow is relevant: asynchrony avoids billed idle request time. Here the evidence instead is a high-volume Lambda function has too much memory allocated for its actual requirements. The selected technical mechanism is Benchmark and tune Lambda memory-duration cost tradeoffs: lambda price depends on memory allocation and execution duration.

Benchmark EC2/ECS continuous compute economics against Lambda solves a different operational problem where a workload has stable continuous high CPU load and Lambda invocation charges exceed EC2. Long steady compute may favor provisioned capacity. It does not directly address a high-volume Lambda function has too much memory allocated for its actual requirements. The case calls for Benchmark and tune Lambda memory-duration cost tradeoffs, since lambda price depends on memory allocation and execution duration.

The competing option Schedule ECS service desired-count reductions with safeguards would be useful when an ECS service's tasks run 24/7 at full size although demand is only daytime. Scaling down idle containers reduces compute charges. The present issue is a high-volume Lambda function has too much memory allocated for its actual requirements; it is more directly addressed by Benchmark and tune Lambda memory-duration cost tradeoffs. Lambda price depends on memory allocation and execution duration.

The competing option Migrate supported short handlers to Lambda with API Gateway would be useful when a low-traffic webhook service runs on an always-on EC2 instance. Serverless per-invocation billing can suit sporadic traffic. The present issue is a Lambda function waits idle on a slow synchronous external request; it is more directly addressed by Make independent long operations asynchronous where business semantics allow. Asynchrony avoids billed idle request time.

The option Benchmark and tune Lambda memory-duration cost tradeoffs addresses a separate condition: a high-volume Lambda function has too much memory allocated for its actual requirements. Lambda price depends on memory allocation and execution duration. The actual requirement is a Lambda function waits idle on a slow synchronous external request. Recommended corrective action: Make independent long operations asynchronous where business semantics allow. Asynchrony avoids billed idle request time.

Schedule ECS service desired-count reductions with safeguards would help if an ECS service's tasks run 24/7 at full size although demand is only daytime. Scaling down idle containers reduces compute charges. That is not the symptom described here: a Lambda function waits idle on a slow synchronous external request. The matching response is Make independent long operations asynchronous where business semantics allow, because asynchrony avoids billed idle request time.

The option Migrate supported short handlers to Lambda with API Gateway addresses a separate condition: a low-traffic webhook service runs on an always-on EC2 instance. Serverless per-invocation billing can suit sporadic traffic. The actual requirement is an ECS service's tasks run 24/7 at full size although demand is only daytime. Recommended corrective action: Schedule ECS service desired-count reductions with safeguards. Scaling down idle containers reduces compute charges.

Benchmark and tune Lambda memory-duration cost tradeoffs would help if a high-volume Lambda function has too much memory allocated for its actual requirements. Lambda price depends on memory allocation and execution duration. That is not the symptom described here: an ECS service's tasks run 24/7 at full size although demand is only daytime. The matching response is Schedule ECS service desired-count reductions with safeguards, because scaling down idle containers reduces compute charges.

For the alternative failure a workload has stable continuous high CPU load and Lambda invocation charges exceed EC2, the control Benchmark EC2/ECS continuous compute economics against Lambda is relevant: long steady compute may favor provisioned capacity. Here the evidence instead is an ECS service's tasks run 24/7 at full size although demand is only daytime. The selected technical mechanism is Schedule ECS service desired-count reductions with safeguards: scaling down idle containers reduces compute charges.

The option Migrate supported short handlers to Lambda with API Gateway addresses a separate condition: a low-traffic webhook service runs on an always-on EC2 instance. Serverless per-invocation billing can suit sporadic traffic. The actual requirement is a workload has stable continuous high CPU load and Lambda invocation charges exceed EC2. Recommended corrective action: Benchmark EC2/ECS continuous compute economics against Lambda. Long steady compute may favor provisioned capacity.

For the alternative failure a high-volume Lambda function has too much memory allocated for its actual requirements, the control Benchmark and tune Lambda memory-duration cost tradeoffs is relevant: lambda price depends on memory allocation and execution duration. Here the evidence instead is a workload has stable continuous high CPU load and Lambda invocation charges exceed EC2. The selected technical mechanism is Benchmark EC2/ECS continuous compute economics against Lambda: long steady compute may favor provisioned capacity.

Schedule ECS service desired-count reductions with safeguards solves a different operational problem where an ECS service's tasks run 24/7 at full size although demand is only daytime. Scaling down idle containers reduces compute charges. It does not directly address a workload has stable continuous high CPU load and Lambda invocation charges exceed EC2. The case calls for Benchmark EC2/ECS continuous compute economics against Lambda, since long steady compute may favor provisioned capacity.