Tracking API usage cost (IoT) using Lambda function

Yes, there are several ways to track and calculate the cost of using IoT devices for multi-tenancy scenarios. Here are some approaches you can consider:

1. Implement a granular tagging strategy: Tag your IoT resources with unique identifiers for each tenant. This allows you to associate costs with specific tenants or groups of devices.

2. Utilize AWS Cost Allocation Tags: Create custom cost allocation tags that align with your tagging strategy. This enables you to segment costs based on tenants or device groups in AWS Cost Explorer.

3. Use AWS Cost Explorer: Leverage Cost Explorer to analyze your IoT costs in detail. You can filter and segment costs based on the tags you've applied, allowing you to isolate expenses for each tenant or device group.

4. Implement custom logging and metrics: Use CloudWatch to set up custom metrics and logs for your IoT devices. This can help you track usage patterns and correlate them with costs.

5. Leverage AWS IoT Core pricing details: AWS IoT Core charges based on the number of messages processed. You can use the MQTT test client to validate the number of messages being handled and compare it with your expected usage.

6. Create a custom Lambda function: Develop a Lambda function that queries the AWS Cost Explorer API to generate detailed cost reports. This function can be scheduled to run periodically using EventBridge, providing regular cost breakdowns for your multi-tenant IoT setup.

7. Monitor specific IoT Core metrics: Use CloudWatch to track metrics such as the number of messages published, number of rules invoked, and number of shadow updates. These metrics can help you understand usage patterns and potential cost drivers.

8. Implement message batching: To optimize costs, consider implementing message batching on your devices to reduce the overall number of transmissions.

9. Use AWS IoT Device Defender: This service can help you monitor device behavior and identify any anomalies that might lead to unexpected costs.

By combining these approaches, you can create a comprehensive system for tracking and calculating IoT device costs in a multi-tenant environment. This will provide you with more granular and real-time cost data compared to relying solely on the daily Cost and Usage Reports from AWS billing.
Sources
monitor Bedrock model charge per invocation resource | AWS re:Post
IoT Core billing way more messages than those published | AWS re:Post
Community | Automate AWS Cost report using EventBridge, Lambda, Cost Explorer API