An organization is designing a globally distributed IoT stream processing application that will ingest telemetry data from smart grid sensors deployed across two geographic regions. The workload is write-heavy, generating over writes per second. You need to design an Azure Cosmos DB for NoSQL solution that ensures write scalability, prevents hot partitions, and maintains high availability even during a regional outage. Which configuration should you recommend?
- Configure the Cosmos DB container with a partition key of deviceId and enable multi-region writes for the Cosmos DB account across the deployed regions.Cevap
- BConfigure the Cosmos DB container with a partition key of region and configure the Cosmos DB account with Locally Redundant Storage (LRS) to handle local replication.
- CConfigure the Cosmos DB container with a partition key of deviceId and authenticate the devices using a Shared Access Signature (SAS) token with a five-year expiration date hardcoded in the device firmware.
- DConfigure the Cosmos DB container with a partition key of deviceType and disable multi-region writes while using Locally Redundant Storage (LRS) to minimize consumption costs.
Cevap
Configure the Cosmos DB container with a partition key of deviceId and enable multi-region writes for the Cosmos DB account across the deployed regions.
The configuration utilizing deviceId as the partition key and enabling multi-region writes is correct because deviceId has high cardinality, which distributes the writes per second evenly across the physical partitions. Multi-region writes provide the necessary high availability and write performance in both regions to survive regional outages.
Adım Adım Çözüm
Anahtar Kavram
Azure Cosmos DB partitioning, multi-region replication, and secure credential design.
Tahmini Süre:1m 30s