Examkin

Question

Difficulty: MediumHigh-Performing and Scalable Storage Solutions

A financial analytics firm is deploying a high-performance computing (HPC) modeling workload on a fleet of Amazon EC2 Linux instances. The application requires a shared, POSIX-compliant storage system that can handle concurrent read and write operations from all instances, deliver sub-millisecond latencies, and scale throughput to at least 80 GB/s80\text{ GB/s} for processing large datasets. Which combination of steps should a solutions architect recommend to meet these requirements? (Select TWO.)

  1. Deploy an Amazon FSx for Lustre file system to store and serve the shared datasets.Answer
  2. Mount the shared file system on the Amazon EC2 instances using the Lustre client.Answer
  3. C
    Configure an Amazon EFS file system with Elastic throughput mode and mount it on the instances.
  4. D
    Provision an Amazon EBS io2 Block Express volume with Multi-Attach enabled and mount it on all instances.
  5. E
    Deploy an Amazon S3 bucket and use AWS Transfer Acceleration to mount it as a local directory on the EC2 instances.

Answer

Deploying an Amazon FSx for Lustre file system and mounting it on the Amazon EC2 instances using the Lustre client.
Amazon FSx for Lustre is a high-performance parallel file system designed specifically for compute-intensive workloads such as machine learning and high-performance computing (HPC). It provides sub-millisecond latencies, supports concurrent access from hundreds of EC2 instances, and scales throughput to hundreds of gigabytes per second. To mount and interact with the parallel file system, the client instances must run the open-source Lustre client.

Step-by-Step Solution

1
Analyze the storage performance and access pattern requirements of the HPC workload.
The workload requires concurrent, POSIX-compliant access from a fleet of hundreds of instances, sub-millisecond latency, and at least 80 GB/s80\text{ GB/s} of throughput scale.
Understanding these requirements is critical to selecting the appropriate AWS storage service.
2
Evaluate FSx for Lustre as the shared file system.
Amazon FSx for Lustre is a parallel file system designed for HPC workloads, supporting concurrent access by hundreds of instances and scaling to hundreds of gigabytes per second of throughput with sub-millisecond latencies.
It matches the scalability and performance requirements of the workload.
3
Determine the connection method for the EC2 fleet to the FSx for Lustre file system.
The open-source Lustre client must be installed on the Linux EC2 instances to mount the shared file system.
This allows the EC2 instances to interact with the file system concurrently with high performance.

Key Concept

HPC storage requirements and the application of Amazon FSx for Lustre to meet high-throughput parallel access demands.
Rate this question

Topics

High-Performing and Scalable Storage Solutions

More questions from this topic