Examkin

Soru

Zorluk: OrtaCost-Optimized Database Design and Capacity Planning

A company runs a critical transactional application on an Amazon RDS for PostgreSQL Multi-AZ DB instance. The database is currently sized as a db.m6g.4xlargedb.m6g.4xlarge instance to handle a combination of transactional writes and a heavy volume of read-only business intelligence (BI) queries. The transactional write workload is steady and low, but the BI queries cause CPU utilization to spike to 90%90\% during business hours. The company wants to optimize costs while maintaining high availability.

Which database design is the most cost-effective solution?

  1. Create an Amazon RDS Read Replica to handle the BI queries, and downsize the primary Multi-AZ DB instance to a db.m6g.largedb.m6g.large instance.Cevap
  2. B
    Configure the application to route the BI queries to the Multi-AZ standby DB instance to utilize its idle capacity, and downsize the primary instance to a db.m6g.largedb.m6g.large instance.
  3. C
    Convert the Multi-AZ deployment to a Single-AZ deployment, create a Read Replica to handle the BI queries, and configure the Read Replica as the primary disaster recovery failover target.
  4. D
    Migrate the database to Amazon DynamoDB using Provisioned Capacity mode to handle both transactional writes and BI queries at a lower cost.

Cevap

Create an Amazon RDS Read Replica to handle the BI queries, and downsize the primary Multi-AZ DB instance to a db.m6g.largedb.m6g.large instance.
The correct option offloads the read-heavy business intelligence queries to an Amazon RDS Read Replica. Since the primary database only needs to handle the steady, low-volume write transactions, the primary Multi-AZ instance (and its standby) can be safely downsized from db.m6g.4xlargedb.m6g.4xlarge to db.m6g.largedb.m6g.large, resulting in significant cost savings while maintaining high availability.

Adım Adım Çözüm

1
Analyze the workload characteristics and database requirements.
The workload has high-volume reads (BI queries) and low-volume writes, requiring high availability (Multi-AZ).
Understanding the separation of read and write traffic is critical to sizing database instances.
2
Select a mechanism to offload read queries.
An Amazon RDS Read Replica is created to handle the BI read queries.
This removes the 90%90\% CPU utilization bottleneck from the primary database instance.
3
Resize the primary Multi-AZ database instance.
Downsize the primary Multi-AZ instance from db.m6g.4xlargedb.m6g.4xlarge to db.m6g.largedb.m6g.large.
Since the primary database now only processes the steady, low-volume write transactions, it no longer needs to be oversized. This reduces the cost of the Multi-AZ deployment.

Anahtar Kavram

Separation of read and write workloads using RDS Read Replicas to optimize Multi-AZ compute costs.
Bu soruyu puanla

Konular

Cost-Optimized Database Design and Capacity Planning