Examkin

Question

Difficulty: HardCost-Optimized Compute Selection and Purchasing Strategies

A media processing company is building a system to process daily video transcoding workloads. The architecture includes a continuous, steady-state orchestration queue worker that runs 24/724/7 on Amazon ECS Fargate, and a highly variable fleet of transcoding workers running containerized tasks to process jobs from an Amazon SQS queue. Each transcoding job takes between 55 to 3030 minutes to complete, can be safely interrupted and retried from the start, and must be completed within 2424 hours of submission. The transcoding tasks retrieve source videos from an Amazon S3 bucket and write the output back to the same bucket. Which combination of purchasing strategies and compute configurations will meet these requirements MOST cost-effectively? (Select TWO.)

  1. Purchase a 33-year Compute Savings Plan to cover the baseline usage of the orchestration queue worker running on Amazon ECS Fargate.Answer
  2. Run the transcoding worker tasks on Amazon ECS Fargate using the Fargate Spot capacity provider.Answer
  3. C
    Purchase a 33-year Compute Savings Plan to cover both the orchestration queue worker and the application's Amazon RDS database instances.
  4. D
    Deploy the transcoding worker logic as AWS Lambda functions configured with maximum timeout to continuously poll and process the SQS queue.
  5. E
    Provision a NAT Gateway in the VPC to route the data transfer of source videos and transcoded files between the private ECS container instances and Amazon S3.

Answer

Purchasing a Compute Savings Plan for the orchestration worker and using Fargate Spot for the transcoding tasks provides the most cost-optimized solution.
The correct combination involves purchasing a Compute Savings Plan to cover the predictable, steady-state 24/724/7 orchestration worker running on ECS Fargate, and running the variable, interruptible video transcoding tasks on ECS Fargate using the Fargate Spot capacity provider to leverage deep discounts.

Step-by-Step Solution

1
Analyze the workload characteristics of the ECS orchestration worker.
The orchestration worker runs continuously (24/724/7) on ECS Fargate, representing a predictable, steady-state baseline compute workload.
Identifying steady-state compute workloads allows the application of commitment discounts like Savings Plans.
2
Determine the most cost-effective purchasing model for the steady-state orchestration worker.
A 33-year Compute Savings Plan is selected because it applies directly to ECS Fargate and offers significant discounts over On-Demand pricing.
Compute Savings Plans are flexible and cover ECS Fargate compute usage, unlike EC2 Instance Savings Plans which only cover EC2.
3
Analyze the workload characteristics of the transcoding workers.
The transcoding workers are highly variable, run containerized tasks, process SQS queue messages, and are highly tolerant of interruptions.
Workloads that are stateless, queue-driven, and interruptible are ideal candidates for Spot capacity.
4
Select the optimal compute and purchasing strategy for the transcoding workers.
Fargate Spot is selected as the capacity provider for these tasks.
Fargate Spot offers spare AWS capacity at steep discounts (up to 70%70\%) compared to On-Demand Fargate, perfectly matching the interruptible nature of the transcoding jobs.

Key Concept

Selecting cost-optimized AWS compute models requires matching steady-state workloads to Savings Plans/Reserved Instances and matching interruptible, flexible workloads to Spot instances.
Rate this question

Topics

Cost-Optimized Compute Selection and Purchasing Strategies