Examkin

Question

Difficulty: HardDatabase and Storage Strategy

A global logistics enterprise is designing a new cloud-native supply chain tracking and routing platform. The platform handles an OLTP workload consisting of real-time package status transitions and route assignments. The system must support up to 80,000 reads/sec80,000 \text{ reads/sec} and 15,000 writes/sec15,000 \text{ writes/sec} during peak hours, with database read latency kept below 15 ms15 \text{ ms}. The disaster recovery requirements specify a Recovery Time Objective (RTO) of less than 10 minutes10 \text{ minutes} and a Recovery Point Objective (RPO) of less than 1 minute1 \text{ minute}. Additionally, an analytics dashboard operating in a separate, dedicated AWS account must be able to securely read data from the database with minimal performance impact on the primary OLTP application. All data must be encrypted at rest. Which database and storage strategy meets these requirements with the lowest operational complexity?

  1. A
    Deploy an Amazon Aurora PostgreSQL Global Database with the primary cluster in the main Region and a secondary cluster in the recovery Region. Encrypt the global database using the default AWS-managed KMS key (aws/rds). Modify the default AWS-managed KMS key policy to grant decrypt permissions to the cross-account analytics role so it can query the secondary database.
  2. Deploy an Amazon Aurora PostgreSQL Global Database with the primary cluster in the main Region and a secondary cluster in the recovery Region. Configure Aurora Auto Scaling for the Aurora Replicas in both regions to handle peak read traffic. Encrypt the database clusters using an AWS KMS customer managed key in each Region, and grant cross-account KMS decrypt and Aurora DB cluster read permissions to the dedicated analytics account.Answer
  3. C
    Deploy a multi-Region Amazon RDS for PostgreSQL database with a Multi-AZ configuration in both the primary and recovery Regions. Scale read capacity to 80,000 reads/sec80,000 \text{ reads/sec} by directing the analytics dashboard and read queries to the Multi-AZ standby instances in each region. Encrypt the databases using customer managed KMS keys.
  4. D
    Deploy a single-Region Amazon Aurora PostgreSQL DB cluster. Configure hourly automated database snapshots and copy them to the recovery Region. Grant the analytics account access to read from the primary DB cluster via a cross-account IAM role. In the event of a regional outage, restore the database from the copied snapshots in the recovery Region.

Answer

Deploy an Amazon Aurora PostgreSQL Global Database encrypted with a customer managed KMS key, utilizing Aurora Auto Scaling for read scaling, and granting cross-account permissions on the key and database resources.
The correct architecture uses Amazon Aurora Global Database to replicate write operations globally with sub-second latency, resolving the RPO requirements. Aurora Replicas handle read scaling to support high-throughput analytics, while Customer Managed Keys (CMKs) enable cross-account access delegation which is impossible using AWS-managed keys.

Step-by-Step Solution

1
Evaluate the database replication and disaster recovery requirement.
Amazon Aurora Global Database provides sub-second data replication lag, easily satisfying the RPO limit of 1 minute. Regional failover can be completed in less than 1 minute, matching the RTO limit of 10 minutes.
Choosing the correct disaster recovery architecture is required to satisfy strict SLA/RTO/RPO limits.
2
Address read scaling for peak OLTP traffic (80,000 reads/sec80,000 \text{ reads/sec}).
Configure Aurora Auto Scaling to dynamically spin up Aurora Replicas to handle the read workload in the primary and secondary clusters.
Aurora Replicas handle read operations without affecting the write performance of the primary instance.
3
Configure encryption and cross-account access for the analytics dashboard.
Use customer managed KMS keys rather than AWS-managed keys to enable modifying the key policy for cross-account access.
AWS-managed keys cannot be shared with external AWS accounts, which is a hard constraint for the analytics workload.

Key Concept

Amazon Aurora Global Database architecture combined with customer managed KMS encryption keys for cross-account access and high performance global read scaling.
Rate this question

Topics

Database and Storage Strategy

More questions from this topic