Examkin

Question

Difficulty: Very hardDatabase and Storage Strategy

An autonomous vehicle technology firm is designing a high-performance simulation platform on AWS. The platform requires a shared file storage solution to support machine learning model training and log analysis.

The workload profiles and performance indicators are as follows:
* Workload Type: Shared File Storage (POSIX-compliant, supporting advisory file locking)
* Active Dataset: 500 TB500\text{ TB} requiring sub-millisecond read latency and up to 12 GB/s12\text{ GB/s} read throughput
* Archive Dataset: 4 PB4\text{ PB} of historical sensor logs that are infrequently accessed but must be immediately available with low latency when a simulation is rerun
* Disaster Recovery Constraints: The primary deployment is in `us-east-1`. A read-only secondary replica must be maintained in `eu-west-1` with a Recovery Point Objective (RPO) of less than 10 minutes10\text{ minutes} and a Recovery Time Objective (RTO) of less than 1 hour1\text{ hour}.
* Security Constraints: All data must be encrypted at rest using Customer Managed Keys (CMKs) in AWS Key Management Service (AWS KMS).

Which storage architecture represents the most cost-effective design that satisfies all latency, throughput, RPO/RTO, and security requirements?

  1. Deploy Amazon FSx for NetApp ONTAP in `us-east-1` using FlexGroup volumes with an 'Auto' tiering policy to transition cold data to capacity pool storage. Encrypt the file system using a regional customer managed KMS key. Deploy a single-AZ FSx for ONTAP file system in `eu-west-1` encrypted with a local customer managed KMS key. Configure NetApp SnapMirror to replicate data from `us-east-1` to `eu-west-1` on a schedule that satisfies the 10-minute10\text{-minute} RPO.Answer
  2. B
    Deploy Amazon FSx for NetApp ONTAP in `us-east-1` using FlexGroup volumes with an 'Auto' tiering policy. Encrypt the file system using the default AWS-managed KMS key (`aws/fsx`). Deploy a single-AZ FSx for ONTAP file system in `eu-west-1` encrypted with the default AWS-managed KMS key (`aws/fsx`) in the destination region. Configure NetApp SnapMirror to replicate data from `us-east-1` to `eu-west-1` on a schedule that satisfies the 10-minute10\text{-minute} RPO.
  3. C
    Deploy Amazon FSx for Lustre (Persistent-2 SSD deployment) in `us-east-1` with data compression enabled. Periodically sync the active training data to Amazon S3. Configure an AWS DataSync task scheduled to run every 12 hours12\text{ hours} to replicate the data from the S3 bucket in `us-east-1` to an S3 bucket in `eu-west-1`, which is attached to a passive FSx for Lustre file system.
  4. D
    Deploy Amazon EFS in `us-east-1` with Provisioned Throughput configured for 12 GB/s12\text{ GB/s} and EFS Lifecycle Management set to transition files to the Infrequent Access (IA) storage class after 14 days14\text{ days}. Enable EFS Replication to copy data to an EFS file system in `eu-west-1`. Encrypt both file systems using customer managed KMS keys.

Answer

Deploy Amazon FSx for NetApp ONTAP in the primary region using FlexGroup volumes with Capacity Pool tiering and customer managed KMS encryption, then replicate to a secondary region cluster using NetApp SnapMirror.
The correct architecture uses Amazon FSx for NetApp ONTAP with FlexGroup volumes to scale throughput to twelve gigabytes per second and support POSIX locking. The Auto tiering policy minimizes storage costs by automatically tiering the four petabytes of cold logs to capacity pool storage. Cross-region replication via NetApp SnapMirror enables block-level replication capable of meeting the sub-ten-minute RPO. Customer Managed Keys are utilized to configure custom KMS key policies allowing cross-region key delegation.

Step-by-Step Solution

1
Analyze performance and scalability constraints.
Amazon FSx for NetApp ONTAP FlexGroup volumes support scaling read throughput to over ten gigabytes per second with sub-millisecond latencies, satisfying the active simulation throughput requirement.
Identifying the file system technology that natively supports the performance scale and POSIX locks is the first step in storage selection.
2
Analyze cost-effective tiering for the archive dataset.
The Auto tiering policy on FSx for ONTAP transparently transitions cold block data to the Capacity Pool, reducing storage costs for the four petabytes of log archives while keeping them immediately accessible.
Storing multiple petabytes of cold logs on high-performance SSD storage is cost-prohibitive, necessitating tiering.
3
Evaluate the disaster recovery replication path.
NetApp SnapMirror replicates metadata and block updates incrementally, easily achieving a recovery point objective under ten minutes, unlike file-based sync methods or high-latency replication schedules.
Ensuring the selected replication mechanism matches the recovery point and recovery time objectives.
4
Evaluate key management and encryption capabilities.
Customer Managed Keys (CMKs) must be utilized because AWS-managed keys do not allow cross-region policy modifications required by NetApp SnapMirror.
Ensuring encryption meets the security policy and allows cross-region cryptographic actions.

Key Concept

Selecting and configuring high-performance shared file storage with cross-region replication and data tiering while satisfying strict compliance and recovery objectives.
Rate this question

Topics

Database and Storage Strategy

More questions from this topic