Examkin

Soru

Zorluk: OrtaCost-Optimized Compute Selection and Purchasing Strategies

A retail company runs a containerized web application on AWS. The application consists of a front-end service hosted on Amazon Elastic Container Service (Amazon ECS) on AWS Fargate that runs 24/7 with a highly predictable baseline load of 8 tasks. The application also runs a nightly backend reporting job on Amazon ECS that processes sales data for 3 hours. The reporting job can tolerate interruptions and resume from checkpoints, but must complete within a 6-hour execution window. The company also stores intermediate reporting logs in Amazon S3 for 10 days before deleting them.

Which two strategies should a solutions architect recommend to minimize compute and storage costs for these workloads? (Select TWO.)

  1. Purchase a Compute Savings Plan to cover the baseline AWS Fargate tasks.Cevap
  2. Configure the Amazon ECS service for the nightly reporting job to use the Fargate Spot capacity provider.Cevap
  3. C
    Purchase a Compute Savings Plan to cover the compute charges for both the AWS Fargate tasks and the Amazon RDS database.
  4. D
    Migrate the nightly reporting job to run on AWS Lambda to eliminate idle compute costs.
  5. E
    Store the intermediate reporting logs in Amazon S3 Standard-Infrequent Access (S3 Standard-IA) and delete them after 10 days.

Cevap

The correct strategies are to purchase a Compute Savings Plan to cover the baseline Fargate tasks, and to configure the Amazon ECS service for the nightly reporting job to use the Fargate Spot capacity provider.
Purchasing a Compute Savings Plan covers the continuous 24/7 baseline of the ECS Fargate tasks, yielding significant discounts compared to On-Demand pricing. Additionally, using Fargate Spot for the nightly reporting job utilizes spare AWS capacity at a heavy discount, which is appropriate because the job is fault-tolerant and has a flexible completion window.

Adım Adım Çözüm

1
Analyze the baseline Fargate tasks workload.
The tasks run 24/7 with a highly predictable baseline load of 8 tasks.
Workloads with stable, predictable, and continuous compute needs are ideal candidates for commitment-based pricing models like Compute Savings Plans to maximize discount rates.
2
Analyze the nightly reporting job workload.
The job runs for 3 hours, can tolerate interruptions, and has a flexible completion window of 6 hours.
Since the job is fault-tolerant and has flexible execution times, Fargate Spot is the most cost-effective compute choice, offering significant discounts by using spare capacity.
3
Evaluate the storage optimization strategy for intermediate logs.
Storing logs in S3 Standard-IA for only 10 days incurs a cost penalty.
Amazon S3 Standard-IA enforces a minimum 30-day storage billing duration. Moving logs that are deleted in 10 days to S3 Standard-IA results in paying for 20 days of non-existent data, making standard S3 more cost-effective.
4
Evaluate the database pricing and scope of Compute Savings Plans.
Compute Savings Plans cannot cover Amazon RDS database instances.
Compute Savings Plans are restricted to EC2, Fargate, and Lambda. They do not apply to database engines or RDS, which require Database Reserved Instances.

Anahtar Kavram

Selecting cost-optimized compute purchasing options (Compute Savings Plans for predictable baseline workloads, Spot instances for fault-tolerant batch workloads) and evaluating minimum storage duration charges in storage tiers.
Bu soruyu puanla

Konular

Cost-Optimized Compute Selection and Purchasing Strategies