Exam AWS Certified Solutions Architect - Associate SAA-C03 topic 1 question 926 discussion

spots might be interrupted and the questions says it has to be available 24/7, rightsizing addresses the cost optimization point but the spot is so risk so I'll go for D

AWS Compute Optimizer now works with Fargate. All your Spot instances could be deleted and your SOL.

❌ Why other options are suboptimal: A. Fargate only + rightsizing via CloudWatch Using Fargate alone for all workloads is more expensive. While rightsizing helps, it doesn’t address bursty traffic cost-effectively. C. Fargate Spot for steady state Fargate Spot is interruptible. It is not suitable for critical steady state workloads that need 24/7 availability. D. Fargate only + AWS Compute Optimizer Compute Optimizer does not yet support Fargate task-level recommendations. Doesn’t solve the cost-efficiency for burst traffic.

B is a better answer

Use Fargate Spot for burst traffic — it’s significantly cheaper (up to 70% savings) ECS Capacity Providers let you define a mix strategy for both types of compute This solution provides: High availability via regular Fargate- Cost savings (via Spot)- Scalability for traffic bursts without manual management why other options are incorrect? A. Only uses standard Fargate with performance tuning tools. Does not address cost optimization or scaling for traffic bursts. No use of Fargate Spot, which misses out on potential savings. C. Using Fargate Spot for steady-state workloads is not reliable. Spot tasks can be interrupted at any time, which is not acceptable for a 24/7 high-availability app. D. AWS Compute Optimizer helps with right-sizing but doesn't handle scaling or burst traffic.

Compute Optimizer can not optimise fargate tasks.

https://repost.aws/knowledge-center/fargate-spot-termination-notice B is correct as you can use ECS task draining to gracefully handle Spot interruptions.

B is the valid answer. Use Fargate for steady-state tasks to ensure high availability and reliability, and Fargate Spot for burst tasks to handle high traffic at a reduced cost. This combination meets the requirements most cost-effectively. D is wrong because rightsizing alone does not optimize for burst traffic scenarios or reduce costs as effectively as incorporating Fargate Spot.

A - It doesn't address the cost optimization for burst traffic. Rightsizing ensures you're not over-provisioning your baseline resources, but it doesn't provide a mechanism to dynamically scale up with cost-effective Spot instances during traffic spikes. B - The application can scale up quickly with Fargate Spot tasks to handle the increased load, and if a Spot task is interrupted, the remaining Fargate tasks can continue to serve traffic (and the system can try to acquire new Spot capacity). C - Fargate Spot is not recommended for steady-state workloads because it can be interrupted. This would compromise the application’s 24/7 availability. D - Like A, it fails to leverage Fargate Spot for cost savings. Combining Fargate for steady state with Fargate Spot for burst traffic, the 24/7 HA is actually guaranteed.

It's D. Compute Optimizer supports recommendations for Fargate Tasks https://aws.amazon.com/about-aws/whats-new/2022/12/aws-compute-optimizer-amazon-ecs-services-aws-fargate/ Option B and C recommend Spot, but that's not an option since the the question asks for 24x7 availability.

D is right answer. Company desire is 24 * 7. So I think spot should be.

available for 24 * 7, spot is out as it can be interrupted any time.

https://aws.amazon.com/blogs/aws-cloud-financial-management/how-to-take-advantage-of-rightsizing-recommendation-preferences-in-compute-optimizer/ be available 24 hours a day, 7 days a week must be highly available => D

This combination leverages the cost benefits of Fargate Spot for burst traffic while ensuring steady performance with regular Fargate instances.

B is the right answer

B- for short work use spot

B is the right choice, the application must be available 24/7