Examkin

Question

Difficulty: HardHigh-Performing and Scalable Storage Solutions

A financial company is designing a high-performance computing (HPC) cluster running risk analysis models on a fleet of Amazon EC2 Linux instances. The workload requires a shared storage solution that supports standard NFSv4 mounts, delivers consistent sub-millisecond latencies for small-file random I/O, scales up to 100,000100\text{,000} IOPS and 3 GB/s3\text{ GB/s} of throughput, and enables the creation of instantaneous, space-efficient writable clones of the datasets for sandbox testing. Which two storage configurations will satisfy these requirements? (Select TWO.)

  1. Deploy an Amazon FSx for OpenZFS file system, mount the volumes on the instances using NFSv4, and use the native ZFS snapshot and cloning capabilities to create staging environments.Answer
  2. Deploy an Amazon FSx for NetApp ONTAP file system, mount the volumes on the instances using NFSv4, and use the NetApp FlexClone feature to create instantaneous writable volume clones.Answer
  3. C
    Deploy an Amazon Elastic File System (Amazon EFS) file system, configure it with Provisioned Throughput, and use AWS Backup to restore copies of the file system to new mount targets for testing.
  4. D
    Deploy an Amazon FSx for Lustre file system, mount the file system on the Linux instances using the NFSv4 protocol, and use Amazon S3 data repository integration to sync and clone datasets.
  5. E
    Deploy an Amazon EBS io2 Block Express volume with Multi-Attach enabled, mount the volume across the fleet, format the volume with the NFSv4 file system, and use EBS snapshots to create clones.

Answer

The correct configurations are Amazon FSx for OpenZFS and Amazon FSx for NetApp ONTAP. Both support NFSv4 client mounts, deliver consistent sub-millisecond latencies, scale to the required performance limits, and offer native, near-instantaneous, space-efficient writable cloning capabilities.
The correct configurations are Amazon FSx for OpenZFS and Amazon FSx for NetApp ONTAP. Both solutions support standard NFSv4 protocols, provide consistent sub-millisecond SSD latencies, comfortably scale to 100,000100\text{,000} IOPS and 3 GB/s3\text{ GB/s} throughput, and natively support near-instantaneous, space-efficient writable cloning of datasets (via native ZFS cloning and NetApp FlexClone respectively).

Step-by-Step Solution

1
Analyze the I/O, latency, protocol, and functional requirements of the shared storage solution.
The requirements are: shared access via standard NFSv4, sub-millisecond latency for small-file random I/O, 100,000100\text{,000} IOPS, 3 GB/s3\text{ GB/s} throughput, and instantaneous space-efficient writable clones.
This establishes the baseline metrics to filter out incompatible AWS storage services.
2
Evaluate Amazon EFS and Amazon EBS Multi-Attach against the requirements.
Amazon EFS cannot guarantee sub-millisecond latency for metadata and random I/O and lacks native instant writable cloning. Amazon EBS Multi-Attach is block storage (cannot natively format with NFSv4 or provide server-side instant writable file system clones).
This eliminates options that violate basic latency, storage type, or cloning speed constraints.
3
Evaluate FSx file systems (Lustre, OpenZFS, ONTAP) against the protocol and performance requirements.
Amazon FSx for Lustre is eliminated because it requires the Lustre client, not standard NFSv4. Amazon FSx for OpenZFS and FSx for NetApp ONTAP both fully support NFSv4, provide sub-millisecond SSD latency, scale to the requested IOPS/throughput, and support instant writable cloning (ZFS clones and NetApp FlexClone).
This identifies the two solutions that perfectly meet all the architectural requirements.

Key Concept

High-performance shared file systems with advanced data management capabilities (snapshots and clones) on AWS.
Rate this question

Topics

High-Performing and Scalable Storage Solutions

More questions from this topic