Medium1 markMultiple Choice
Domain 3.5: Cost OptimizationCost OptimizationEMRSpot Instances

AWS SAP-C02 · Question 30 · Domain 3.5: Cost Optimization

A data engineering team is using Amazon EMR to process large datasets. The clusters are currently running on On-Demand EC2 instances. The workloads are fault-tolerant, flexible in their completion time, and run periodically throughout the week. The team wants to significantly reduce compute costs without changing the application code. Which strategy is MOST cost-effective?

Answer options:

A.

Use Amazon EC2 Spot Instances for the EMR task nodes and On-Demand instances for the core and master nodes.

B.

Purchase EC2 Instance Savings Plans for all nodes in the EMR cluster.

C.

Use Spot Instances for the master, core, and task nodes.

D.

Migrate the workloads to AWS Lambda.

How to approach this question

Understand the roles of EMR nodes (Master, Core, Task) and which are safe to run on interruptible Spot instances.

Full Answer

A.Use Amazon EC2 Spot Instances for the EMR task nodes and On-Demand instances for the core and master nodes.✓ Correct
Use Amazon EC2 Spot Instances for the EMR task nodes and On-Demand instances for the core and master nodes.
Amazon EMR clusters consist of Master, Core, and Task nodes. Master nodes manage the cluster. Core nodes run tasks and store data in HDFS. Task nodes only run tasks and do not store HDFS data. Because the workload is fault-tolerant, you can use highly discounted Spot Instances for the Task nodes. If a Spot instance is reclaimed, EMR simply provisions a new one and retries the task.

Common mistakes

Putting Core or Master nodes on Spot instances, which can corrupt the cluster.

Practice the full AWS Solutions Architect Professional SAP-C02 Practice Exam 5

75 questions · hints · full answers · grading

More questions from this exam