Medium1 markMultiple Choice
Domain 2.5: PerformanceComputeStorageCost OptimizationHPC

AWS SAP-C02 · Question 71 · Domain 2.5: Performance

A media company is migrating its video rendering workload to AWS. The workload is highly parallelizable, fault-tolerant, and requires massive amounts of compute power for short bursts. The company wants to minimize compute costs. They also need a shared file system to store the raw video files that provides high throughput and can be accessed by thousands of EC2 instances simultaneously. Which combination of services should the Architect choose? (Select TWO)

Answer options:

A.

Use Amazon EC2 Spot Instances managed by AWS Batch for the compute workload.

B.

Use Amazon FSx for Lustre for the shared file system.

C.

Use Amazon EC2 Reserved Instances for the compute workload.

D.

Use Amazon Elastic File System (EFS) for the shared file system.

E.

Use AWS Lambda for the video rendering compute workload.

F.

Use Amazon S3 as the direct shared file system mounted via S3 File Gateway.

How to approach this question

Match the fault-tolerant compute requirement to Spot Instances, and the high-throughput parallel file system requirement to FSx for Lustre.

Full Answer

Use Amazon EC2 Spot Instances managed by AWS Batch for the compute workload., Use Amazon FSx for Lustre for the shared file system.
For fault-tolerant, bursty workloads, Amazon EC2 Spot Instances offer the lowest cost. AWS Batch can manage the Spot fleet, automatically retrying jobs if an instance is reclaimed. For the storage layer, video rendering is a High Performance Computing (HPC) workload. Amazon FSx for Lustre is specifically designed for HPC, providing sub-millisecond latencies, up to hundreds of gigabytes per second of throughput, and millions of IOPS.

Common mistakes

Choosing EFS instead of FSx for Lustre for HPC workloads.

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

75 questions · hints · full answers · grading

More questions from this exam