Examkin

Question

Difficulty: MediumResilient Database Configurations and High Availability

A media streaming company is launching an on-demand video platform on AWS. The application database workload is highly unpredictable, experiencing sudden surges of read and write queries during popular content releases, followed by extended periods of low activity. The company requires a database design that supports high availability within a single AWS Region with a Recovery Time Objective (RTO) of less than 30 seconds and a Recovery Point Objective (RPO) of 0 (no data loss) during an Availability Zone outage. Additionally, the database must dynamically scale compute capacity up and down to match workload demands without manual intervention or connection disruption, while allowing read workloads to scale independently. Which database configuration will meet these requirements?

  1. Deploy an Amazon Aurora PostgreSQL Serverless v2 DB cluster with a writer instance and a reader instance in different Availability Zones, and configure the application to use the cluster endpoint for writes and the reader endpoint for reads.Answer
  2. B
    Deploy an Amazon RDS for PostgreSQL DB instance in a single Availability Zone, and configure a Read Replica in a different Availability Zone. Configure the application to direct writes to the primary instance, and design a custom failover script to promote the Read Replica during an outage.
  3. C
    Deploy a single-AZ Amazon Aurora DB cluster, and set up a separate standby Aurora DB cluster in another Availability Zone. Configure Amazon Route 53 latency routing to automatically failover and route database writes between the two clusters during an outage.
  4. D
    Deploy an Amazon RDS for PostgreSQL DB instance in a Multi-AZ deployment. Set up a second RDS DB instance as a scaled-down warm standby in a different AWS Region, and configure the application to failover to the regional standby within 30 seconds.

Answer

Deploy an Amazon Aurora PostgreSQL Serverless v2 DB cluster with a writer instance and a reader instance in different Availability Zones, and configure the application to use the cluster endpoint for writes and the reader endpoint for reads.
The correct configuration utilizes Amazon Aurora Serverless v2 in a Multi-AZ deployment with both a writer and a reader instance. Aurora Serverless v2 automatically scales compute capacity up and down based on application demand without causing connection disruption. By having a reader instance in a separate Availability Zone, Aurora can automatically fail over to the reader in less than 30 seconds (meeting the RTO constraint). Because both instances share the same distributed Aurora storage volume across multiple Availability Zones, no data loss occurs during a failover (meeting the RPO of 0). The reader endpoint allows the application to route read queries to the reader instance, scaling read capacity independently.

Step-by-Step Solution

1
Identify the RTO, RPO, and compute scaling constraints from the scenario.
The solution requires high availability with RTO < 30 seconds, RPO = 0 (no data loss) in a single Region, dynamic compute scaling without connection drops, and read scaling.
Understanding these metrics eliminates disaster recovery options (which have RPO > 0 due to asynchronous cross-region replication) and standard RDS scaling options (which require downtime to change instance classes).
2
Evaluate the replication and failover mechanisms of the database engines.
Amazon Aurora replicates data across three Availability Zones using a shared storage volume, ensuring RPO = 0. Failover to an Aurora replica takes less than 30 seconds.
This guarantees the required HA and recovery objectives are met automatically during an Availability Zone outage.
3
Determine the scaling mechanism that prevents connection disruptions.
Amazon Aurora Serverless v2 scales compute capacity dynamically in place, preventing connection drops during workload spikes.
This satisfies the requirement for seamless scaling under unpredictable workloads.

Key Concept

Amazon Aurora Serverless v2 Multi-AZ deployments with reader instances provide high availability (RTO < 30 seconds), zero data loss (RPO = 0), and instant compute scaling without connection disruption.
Estimated Time:2m 0s
Rate this question

Topics

Resilient Database Configurations and High Availability

More questions from this topic