Examkin

Question

Difficulty: HardDatabase and Storage Strategy

A medical technology company is designing a new patient monitoring system that registers real-time telemetry from wearable devices (OLTP workload). The system must scale to ingest up to 100,000100,000 writes per second. The telemetry data must be buffered and cached for real-time visualization with less than 10 ms10\text{ ms} read latency. The cache requires high availability, automated failover, and data persistence to survive node failures. The historical telemetry data must be persisted in a database, and the architecture must ensure horizontal scaling of read operations during peak diagnostic hours. Finally, the company requires a disaster recovery strategy to a secondary AWS Region with a Recovery Time Objective (RTO) of less than 5 minutes5\text{ minutes} and a Recovery Point Objective (RPO) of less than 1 minute1\text{ minute}. Which of the following database and storage strategies meets these requirements with the lowest operational complexity?

  1. A
    Deploy Amazon ElastiCache for Redis with Multi-AZ and auto-failover enabled to serve the caching layer, configuring database persistence. Deploy a single-region Amazon Aurora PostgreSQL cluster as the persistent database engine. Configure AWS Backup to copy daily snapshots to the disaster recovery region, and implement an automated CloudFormation template to restore the database to a new Aurora cluster in the secondary region during a failover event.
  2. B
    Deploy Amazon ElastiCache for Memcached to serve the caching layer. Deploy Amazon Aurora Global Database as the persistent database engine. Use Aurora Replicas in the primary AWS Region with Application Auto Scaling to handle dynamic read scaling, and provision a secondary Aurora cluster in the disaster recovery region.
  3. Deploy Amazon ElastiCache for Redis with Multi-AZ and auto-failover enabled to serve the caching layer, configuring database persistence. Deploy Amazon Aurora Global Database as the persistent database engine. Use Aurora Replicas in the primary AWS Region with Application Auto Scaling to handle dynamic read scaling, and provision a secondary Aurora cluster in the disaster recovery region.Answer
  4. D
    Deploy Amazon ElastiCache for Redis with Multi-AZ and auto-failover enabled to serve the caching layer, configuring database persistence. Deploy Amazon RDS for PostgreSQL in a Multi-AZ deployment as the primary database. Configure the application to route read queries directly to the Multi-AZ standby instance during peak hours to achieve horizontal read scaling, and deploy a cross-region read replica in the disaster recovery region.

Answer

Deploy Amazon ElastiCache for Redis with Multi-AZ and auto-failover enabled for caching, and utilize Amazon Aurora Global Database for persistence with auto-scaled Aurora Replicas and a secondary cluster in the disaster recovery region.
The correct strategy combines Amazon ElastiCache for Redis and Amazon Aurora Global Database. ElastiCache for Redis supports the required sub-millisecond latencies, Multi-AZ automated failover, and data persistence to survive node losses. Amazon Aurora Global Database facilitates sub-second cross-region replication to satisfy the sub-minute RPO, while supporting rapid promotion to meet the sub-5-minute RTO. Aurora Replicas in the primary region scale horizontally and automatically using Application Auto Scaling to accommodate variable read loads.

Step-by-Step Solution

1
Analyze the caching requirements (latency, high availability, failover, and persistence).
Identify that Amazon ElastiCache for Redis is required because it supports multi-AZ configurations with auto-failover and data persistence mechanisms (such as AOF or RDB snapshots), whereas Memcached does not support persistence or replication.
This filters out strategies utilizing Memcached.
2
Evaluate the database options for read scalability.
Determine that Amazon Aurora Replicas with Application Auto Scaling can horizontally scale read workloads dynamically. In contrast, Amazon RDS Multi-AZ standby instances are passive and cannot be targeted for read queries.
This eliminates solutions attempting to route read traffic directly to an RDS Multi-AZ standby instance.
3
Verify the disaster recovery parameters (RTO < 5 minutes, RPO < 1 minute).
Amazon Aurora Global Database uses dedicated, low-latency replication that yields a typical RPO of less than 1 second and allows promotion of the secondary cluster in less than 5 minutes. Daily backup copying strategies cannot achieve sub-minute RPO.
This confirms that Aurora Global Database is the only design that fulfills the RPO and RTO bounds with low operational overhead.

Key Concept

Selecting high-performance cache and database layers that support multi-AZ persistence, horizontal read scaling, and low RTO/RPO cross-region replication.
Estimated Time:3m 0s
Rate this question

Topics

Database and Storage Strategy

More questions from this topic