AZ-305 Question 141
Select 2A global logistics company needs to store real-time shipping logs (semi-structured data) and scanned shipping labels (unstructured data) in Azure. They must ensure continuous availability with minimal downtime if an entire region fails. Which two solutions should you recommend to provide high availability for both data types?
- A
Deploy Azure Cosmos DB across multiple write regions for shipping logs and configure Azure Blob Storage with RA-GZRS for scanned labels.
- B
Deploy a single-region Azure SQL Database for shipping logs with automatic failover group, and use standard LRS for scanned labels.
- C
Deploy Azure Cosmos DB in a single region for shipping logs with autoscale throughput, and store scanned labels in Azure Blob Storage using standard GRS.
- D
Deploy Azure Storage with ZRS for shipping logs in Table Storage, and GRS for scanned labels in Blob Storage.
- E
Deploy Azure Cosmos DB with multi-region read replicas for shipping logs, plus Azure Blob Storage with RA-GRS for scanned labels.
Show answer and explanation
Correct answers: A, E
Explanation
When designing for high availability of both semi-structured and unstructured data across regions, you must ensure each data store is replicated to at least one secondary region, with the ability to fail over automatically or provide read access in the event of a primary region outage. Azure Cosmos DB with multi-region writes or read replicas meets enterprise SLAs for region-level resilience. For unstructured data, Azure Blob Storage with RA-GRS or RA-GZRS provides replication to a secondary region and read access if the primary region is unavailable. Refer to Azure Cosmos DB documentation (https://learn.microsoft.com/azure/cosmos-db/high-availability) and Azure Storage redundancy guidance (https://learn.microsoft.com/azure/storage/common/storage-redundancy) for more details on high availability design.
- A. Correct.
Option 1 is CORRECT. Multi-region writes in Azure Cosmos DB provide immediate failover with minimal downtime for semi-structured data (logs). RA-GZRS (Read-Access Geo-Zone-Redundant Storage) for Blob Storage ensures zonal redundancy within the primary region and geo-redundancy to a secondary region with read-access in case of a region outage.
- B. Incorrect.
Option 2 is INCORRECT. While a failover group can help with some forms of disaster recovery for a SQL Database, a single-region deployment for shipping logs does not address region-level failure properly by itself. In addition, using only LRS for the scanned labels does not provide cross-region redundancy.
- C. Incorrect.
Option 3 is INCORRECT. A single-region Cosmos DB with autoscale throughput is good for scaling, but it does not provide cross-region high availability. Standard GRS replicates data to a secondary region, but it does not allow direct read access from the secondary region if the primary region fails.
- D. Incorrect.
Option 4 is INCORRECT. ZRS (Zone-Redundant Storage) only replicates data across zones in the same region, which does not protect against a full regional outage. GRS for unstructured data does replicate across regions, but you still lack a multi-region HA solution for the shipping logs in Table Storage.
- E. Correct.
Option 5 is CORRECT. Multi-region read replicas in Azure Cosmos DB allow near real-time replication and failover for shipping logs. RA-GRS (Read-Access Geo-Redundant Storage) supports automatic replication of Blob data to a secondary region and provides read access when the primary is unavailable.