A medical imaging enterprise hosts a PACS (Picture Archiving and Communication System) on … — AWS Certified Solutions Architect - Associate

A medical imaging enterprise hosts a PACS (Picture Archiving and Communication System) on AWS. The application runs on a fleet of auto-scaled Amazon EC2 instances across multiple Availability Zones in the primary Region. The architecture requires:
- An Amazon RDS for PostgreSQL database hosting clinical transaction metadata.
- A shared storage layer containing raw DICOM image files (average size $12 \text{ MB}$ ) that must be read and written concurrently by the EC2 fleet.

The enterprise must design a disaster recovery (DR) architecture in a secondary AWS Region. The DR plan requires a Recovery Point Objective (RPO) of under $15 \text{ minutes}$ and a Recovery Time Objective (RTO) of under $10 \text{ minutes}$ for both the database and the image files.

Which architectural design meets these requirements while minimizing recovery time?

Deploy the primary database in a Multi-AZ configuration, and create a cross-Region read replica in the DR Region to be promoted upon failover. Store the DICOM images on a Regional Amazon Elastic File System (Amazon EFS) file system, and configure Amazon EFS Replication to replicate the file system to the DR Region. Deploy a Warm Standby environment in the DR Region with a pre-provisioned, scaled-down EC2 fleet.Cevap
B
Deploy the primary database in a Multi-AZ configuration, and create a cross-Region read replica in the DR Region. Store the DICOM images on a Regional Amazon EFS file system, and use an AWS Backup lifecycle policy to transition and replicate the EFS backups to an Amazon S3 Glacier Flexible Retrieval vault in the DR Region. Deploy a Warm Standby environment in the DR Region.
C
Deploy the primary database in a Multi-AZ configuration, and create a cross-Region read replica in the DR Region. Store the DICOM images on a Regional Amazon EFS file system, and configure Amazon EFS Replication to the DR Region. Deploy a Pilot Light environment in the DR Region with all EC2 application instances stopped, relying on automated scripts to start the instances immediately upon failover.
D
Deploy a Single-AZ primary database with a local read replica in a different Availability Zone configured as the primary automatic failover target. Store the DICOM images on Amazon EBS General Purpose SSD (gp3) volumes attached to the EC2 instances, and configure EBS snapshot replication to the DR Region. Deploy a Warm Standby environment in the DR Region.

Cevap

Deploy the primary database in a Multi-AZ configuration with a cross-Region read replica, store the diagnostic files on Regional Amazon EFS with cross-Region replication enabled, and deploy a Warm Standby environment in the secondary region.

The correct solution addresses high availability at the database layer (Multi-AZ) and provides cross-Region disaster recovery using RDS cross-Region Read Replicas. For storage, Regional Amazon EFS enables simultaneous read/write access across multiple Availability Zones, and EFS Replication fulfills the disaster recovery requirement asynchronously with a very low RPO. Deploying a Warm Standby (with pre-provisioned, scaled-down EC2 instances) ensures the systems can scale up to full production capacity within the

10 \text{ minute}

RTO.

Adım Adım Çözüm

Determine the shared storage requirements for concurrent access and high availability.

Amazon EFS Regional storage is selected because it allows concurrent read/write access from EC2 instances across multiple Availability Zones with sub-millisecond local latencies.

The EC2 application fleet must access the same files concurrently across different AZs, which is supported by EFS but not by EBS gp3 volumes.

Determine the replication strategy to achieve the

15 \text{ minute}

RPO in the DR Region.

Amazon EFS Replication is configured to replicate files asynchronously to the destination Region, and a cross-Region RDS Read Replica is created for the metadata database.

EFS Replication typically replicates changes within minutes, and cross-Region read replicas provide continuous asynchronous updates to the secondary database, satisfying the low RPO.

Select the appropriate disaster recovery environment strategy to meet the

10 \text{ minute}

RTO.

A Warm Standby strategy is implemented in the secondary Region, keeping a minimally scaled, running EC2 fleet ready to receive traffic.

A Warm Standby ensures that instances are already running and can be scaled out rapidly, meeting the sub-

10 \text{ minute}

RTO, whereas a Pilot Light setup with stopped instances would introduce startup delays that violate the RTO.

Anahtar Kavram

Designing a resilient and highly available multi-region architecture using Amazon EFS replication, Multi-AZ database configurations, and a Warm Standby DR strategy to meet strict RTO and RPO requirements.

Tahmini Süre:1m 30s

Soru

Cevap

Adım Adım Çözüm

Anahtar Kavram

Sık Yapılan Hatalar

Konular

Bu konudan daha fazla soru