Examkin

Soru

Zorluk: OrtaDesign Relational Database HA and DR

A company is designing a high availability (HA) and disaster recovery (DR) solution for an on-premises relational database that is being migrated to Azure. The database workload has the following characteristics:
- It requires SQL Server Agent jobs for nightly data maintenance.
- It relies on cross-database queries between three distinct databases.

The DR design must satisfy the following requirements:
- In the event of a regional outage, failover to the secondary region must occur automatically with minimal data loss.
- The application connection strings must remain unchanged after a failover.
- Read-heavy analytics queries must be offloaded to the secondary region to avoid impacting transaction processing.

Which two configurations should you include in the design to meet the requirements? (Select two.)

  1. Azure SQL Managed Instance configured within an auto-failover group across two regions, using the read-write listener for transactions and the read-only listener for analyticsCevap
  2. Global virtual network peering between the delegated subnets of the SQL Managed Instances in both regions to facilitate database replication trafficCevap
  3. C
    An Azure SQL Database elastic pool configured with an auto-failover group to host the databases and manage the automatic failover
  4. D
    Azure SQL Managed Instance configured with active geo-replication for the individual databases, using manual failover orchestration

Cevap

Azure SQL Managed Instance configured within an auto-failover group across two regions, using the read-write listener for transactions and the read-only listener for analytics, along with global virtual network peering between the delegated subnets of the SQL Managed Instances in both regions.
To support SQL Server Agent and cross-database queries, Azure SQL Managed Instance must be selected. To support automatic failover and avoid connection string changes, an auto-failover group must be configured. The auto-failover group provides a read-write listener for transactional workloads and a read-only listener to offload read-heavy analytics to the secondary instance. For the Managed Instances to replicate data, global virtual network peering must be established between the delegated subnets in both regions.

Adım Adım Çözüm

1
Analyze the application compatibility requirements.
Since the workload requires SQL Server Agent and cross-database queries, Azure SQL Database (Single/Elastic Pool) is ruled out. Azure SQL Managed Instance is selected as the target database service.
Azure SQL Managed Instance provides near 100% compatibility with on-premises SQL Server features like SQL Agent and cross-database queries.
2
Determine the disaster recovery mechanism.
Auto-failover groups must be configured. They support automatic failover, database-level replication, and provide read-write and read-only listener endpoints.
This configuration meets the requirement for automatic failover without changing application connection strings, and allows routing read-heavy workloads to the secondary instance.
3
Determine the network connectivity requirements.
Configure virtual network peering between the virtual networks containing the delegated subnets of the primary and secondary SQL Managed Instances.
SQL Managed Instance failover groups require direct IP connectivity between the instances, which is established using global virtual network peering or VPN gateways.

Anahtar Kavram

Designing high availability and disaster recovery for relational databases on Azure using SQL Managed Instance auto-failover groups and networking prerequisites.
Bu soruyu puanla

Konular

Design Relational Database HA and DR

Bu konudan daha fazla soru