A financial analytics firm is designing a high-performance computing (HPC) pipeline on AWS… — AWS Certified Solutions Architect - Associate

A financial analytics firm is designing a high-performance computing (HPC) pipeline on AWS to process market datasets. The pipeline runs on a cluster of 50 Linux-based Amazon EC2 instances that require concurrent, shared access to a file system. The storage solution must deliver sub-millisecond latencies, support throughput up to $12\text{ GB/s}$ , and integrate directly with an Amazon S3 data lake to load and write back data. Which two AWS storage configurations will satisfy these requirements? (Select two.)

Create an Amazon FSx for Lustre file system using SSD storage and configure a data repository association linked to the Amazon S3 bucket.Answer
Select the Persistent SSD deployment type for the FSx for Lustre file system to achieve the required throughput and latency.Answer
C
Provision an Amazon EBS gp3 volume and enable Multi-Attach to share the volume across the EC2 instances.
D
Create an Amazon EFS file system configured with Elastic throughput and mount it on all 50 EC2 instances.
E
Deploy an Amazon FSx for Windows File Server file system with HDD storage linked to the S3 bucket.

Answer

Create an Amazon FSx for Lustre file system using SSD storage with a data repository association linked to the Amazon S3 bucket, and select the Persistent SSD deployment type for the file system.

The correct options are to create an Amazon FSx for Lustre file system using SSD storage with a data repository association linked to the Amazon S3 bucket, and select the Persistent SSD deployment type. FSx for Lustre is designed for high-performance computing (HPC) workloads, offering sub-millisecond latencies and massive throughput. The Persistent SSD deployment type provides high-throughput durable storage, while the data repository association enables seamless integration with the Amazon S3 data lake.

Step-by-Step Solution

Evaluate the workload requirements.

Identify that the system requires shared concurrent access for 50 instances, sub-millisecond latencies,

12\text{ GB/s}

throughput, and native integration with Amazon S3.

This establishes the constraints needed to choose between EBS, EFS, and FSx solutions.

Eliminate incompatible storage systems.

EBS gp3 is ruled out because it lacks Multi-Attach support and is limited to 16 instances. EFS is ruled out due to higher typical latencies (low millisecond) and lack of native data repository integration with S3. FSx for Windows File Server is ruled out due to OS incompatibility and insufficient performance of HDD storage.

Eliminating unsuitable technologies narrows the choices to FSx for Lustre.

Select the correct FSx for Lustre configurations.

Choose Amazon FSx for Lustre with SSD storage and a data repository association to link to S3. Select the Persistent SSD deployment type to provide durable, low-latency, and high-throughput storage.

This configuration meets all latency, throughput, scale, and integration requirements.

Key Concept

High-Performance Storage with FSx for Lustre and S3 Integration

Question

Answer

Step-by-Step Solution

Key Concept

Common Mistakes

Topics

More questions from this topic