Examkin

Question

Difficulty: MediumCost-Optimized Compute Selection and Purchasing Strategies

A financial services firm runs a risk analysis platform on AWS. The platform has a steady-state baseline workload consisting of containerized microservices running on Amazon ECS with AWS Fargate. Every night, the firm runs a memory-intensive batch processing job on Amazon EC2 that takes exactly 3 hours to complete. The batch processing application is resilient and can save its state, allowing it to resume if interrupted. During execution, the batch job writes large amounts of temporary intermediate data to Amazon S3, which is deleted automatically after 5 days. The database layer consists of an Amazon RDS for PostgreSQL database instance. Which two actions should a solutions architect recommend to optimize compute and storage costs? (Select TWO.)

  1. Purchase a Compute Savings Plan to cover the baseline Fargate container workloads.Answer
  2. Launch the nightly batch processing jobs using EC2 Spot Instances.Answer
  3. C
    Purchase a Compute Savings Plan to cover the Amazon RDS for PostgreSQL database instance.
  4. D
    Configure an Amazon S3 Lifecycle rule to transition the temporary intermediate data to S3 Standard-Infrequent Access (S3 Standard-IA) immediately upon creation.
  5. E
    Migrate the 3-hour nightly batch processing job to AWS Lambda to pay only for the exact duration of the execution.

Answer

Purchase a Compute Savings Plan to cover the baseline Fargate container workloads, and launch the nightly batch processing jobs using EC2 Spot Instances.
Purchasing a Compute Savings Plan provides the most cost-effective coverage for the baseline Fargate microservices workload since Compute Savings Plans apply automatically to Fargate. Using EC2 Spot Instances is the most cost-optimized strategy for the nightly batch job because the workload is flexible and resilient to interruptions, allowing the firm to take advantage of the steep discounts offered by Spot pricing.

Step-by-Step Solution

1
Analyze the baseline workload and apply the correct purchasing model.
Identify that the steady-state baseline containers run on AWS Fargate. Select a Compute Savings Plan, which covers Fargate and offers significant discounts for consistent usage.
Compute Savings Plans are flexible and automatically apply to Fargate, whereas EC2 Instance Savings Plans or EC2 Reserved Instances do not.
2
Analyze the batch processing workload properties.
Identify that the nightly batch job runs on EC2, is memory-intensive, runs for 3 hours, and is resilient to interruptions.
Since the workload is fault-tolerant and interruptible, EC2 Spot Instances are the most cost-effective option, offering up to 90% savings over On-Demand pricing.
3
Evaluate the storage requirements for the temporary batch data.
Determine that the intermediate data is deleted after 5 days. Keep the data in S3 Standard rather than transitioning it to S3 Standard-IA.
S3 Standard-IA enforces a 30-day minimum storage charge. Deleting data after 5 days in S3 Standard-IA incurs a minimum duration fee, neutralizing any storage class savings.

Key Concept

Selecting cost-optimized compute purchasing strategies (Savings Plans and Spot Instances) while avoiding scope misapplications and storage duration penalties.
Rate this question

Topics

Cost-Optimized Compute Selection and Purchasing Strategies