Examkin

Question

Difficulty: HardDatabase High Availability

A SysOps Administrator is managing a write-heavy PostgreSQL database on Amazon RDS. The database is currently deployed as a standard Multi-AZ DB instance. During a recent Availability Zone outage, the database failed over to the standby instance, but the application experienced approximately 80 seconds of downtime. The administrator needs to migrate the database to a configuration that reduces the failover time to under 35 seconds while still allowing the standby instances to serve read traffic. Which database configuration and architectural transition should the administrator implement to meet these requirements?

  1. Migrate the database to an Amazon RDS Multi-AZ DB cluster. This configuration uses semi-synchronous replication to two readable standby instances in different Availability Zones, reducing failover time because the standby instances are already running and do not require database crash recovery.Answer
  2. B
    Create an Amazon RDS Read Replica in a different Availability Zone and configure an Amazon Route 53 failover routing policy to automatically promote the replica and route write traffic to it during an outage.
  3. C
    Configure an Amazon Route 53 latency-based routing policy to split write traffic between the primary DB instance and a read replica, enabling automatic failover if the primary instance becomes unhealthy.
  4. D
    Modify the existing Multi-AZ DB instance to run in an active-active configuration across two Availability Zones, and configure a Network Load Balancer to route traffic to the active instance.

Answer

Migrate the database to an Amazon RDS Multi-AZ DB cluster. This configuration uses semi-synchronous replication to two readable standby instances in different Availability Zones, reducing failover time because the standby instances are already running and do not require database crash recovery.
Migrating to an Amazon RDS Multi-AZ DB cluster is the correct solution. Unlike standard Multi-AZ DB instances which have a passive standby that requires database recovery during failover, a Multi-AZ DB cluster maintains two active reader instances in different Availability Zones. Because these instances are already running and serving read traffic, failover is faster (typically under 35 seconds) and requires no database recovery stage.

Step-by-Step Solution

1
Evaluate the limitations of the current standard Multi-AZ DB instance deployment.
The standard Multi-AZ DB instance uses synchronous replication at the storage level to a single passive standby, which requires database engine startup and recovery during failover, leading to failover times of 1-2 minutes.
Understanding the baseline performance is necessary to identify why the current architecture fails to meet the target sub-35-second failover requirement.
2
Identify the high availability database options in Amazon RDS that support readable standbys and faster failover.
An Amazon RDS Multi-AZ DB cluster deployment provides one writer and two readable reader DB instances across three Availability Zones.
This deployment option is designed specifically for lower failover times (typically under 35 seconds) and read scaling from the standbys.
3
Determine the replication and failover mechanism of the Multi-AZ DB cluster.
The DB cluster uses semi-synchronous engine-level replication. Because the standby reader instances are already active and running, they do not need crash recovery upon failover, allowing rapid promotion to writer status.
Confirming the technical mechanism ensures that the chosen option meets both the failover time reduction and the read-traffic serving requirement.

Key Concept

Amazon RDS Multi-AZ DB Cluster Architecture and Failover Dynamics
Rate this question

Topics

Database High Availability

More questions from this topic