A developer is building a backend service for an IoT-based smart agricultural monitoring system. The system receives real-time telemetry from sensor nodes deployed in the field. Each sensor node publishes status updates containing soil moisture and temperature readings. The serialized payload size for each status update is , and the application receives and writes status updates per second directly to an Amazon DynamoDB table. A management dashboard displays a summary of the last telemetry updates for a specific sensor node by making requests per second using eventually consistent reads. Which two of the following actions should the developer perform to design the database and client access layer efficiently?
- Provision Write Capacity Units (WCUs) for the table to handle the sensor telemetry ingestion.Cevap
- Query the table using a KeyConditionExpression on the sensor ID partition key, and provision Read Capacity Units (RCUs) to support the dashboard workload.Cevap
- CScan the table using a FilterExpression on the sensor ID, and provision Read Capacity Units (RCUs) to retrieve the dashboard telemetry.
- DInitialize the Amazon DynamoDB client in the application by hardcoding the AWS Access Key ID and Secret Access Key of a dedicated IAM user to bypass IAM role evaluation latency.
- EProvision Write Capacity Units (WCUs) for the table, calculating the write capacity as without rounding up individual items.
Cevap
Provision Write Capacity Units (WCUs) for the table and query the table using a KeyConditionExpression on the sensor ID partition key, provisioning Read Capacity Units (RCUs).
The correct options are the ones recommending provisioning WCUs for table writes and using the Query API with RCUs. For the write capacity, updates per second at (rounded up to per item) requires WCUs. For the read capacity, querying items totaling (rounded up to ) at eventually consistent queries per second requires RCUs ().
Adım Adım Çözüm
Anahtar Kavram
Calculating DynamoDB read and write capacity units for item-level writes and Query operations under specific consistency models, combined with secure client initialization.