AWS Carbon Report

Bugs, suggestions and feedback

Bug reports, suggestions and feedback to GitHub please!

1. Enable Carbon Data Exports to S3

• Navigate to the Billing and Cost Management console.
• In the left sidebar, select Data Exports.
• Click Create and choose Carbon emissions as the data table.
• Configure the Export:
• • Export name: e.g., carbon_emissions_monthly.
• • S3 bucket: Choose an existing bucket or create a new one.
• • Format: Select Parquet. (It’s much faster and cheaper to query in Athena than CSV).
• • Compression: Snappy.
• • Versioning: Select "Overwrite existing data export file" if you want a clean "latest" state, or "Create new export version" for historical tracking.

[!IMPORTANT] The 3-Month Lag: AWS carbon data is updated monthly but typically has a 3-month delay (e.g., in June, you'll see data for March). Your Grafana dashboard will look empty if you filter for "Last 7 days."

2. Set Up the Athena Table Once the first export lands in S3 (this can take up to 24 hours), you need to tell Athena how to read it.

AWS provides a schema for this, but the easiest way is to use an AWS Glue Crawler pointed at your S3 export path. Alternatively, you can run a manual CREATE TABLE script in the Athena console.

Pro-Tip: If you use the Glue Crawler, it will automatically handle the partitions (like usage_period=2025-01) that AWS creates.

3. Connect Grafana to Athena Use the Athena Datasource Plugin for Grafana, the setup is straightforward:

• IAM Permissions: Ensure the IAM role/user Grafana is using has:
• • athena:StartQueryExecution
• • s3:GetObject (on your Carbon export bucket)
• • glue:GetTable / glue:GetPartitions
• Configure Datasource: In Grafana, point to your Athena workgroup and the database where you created the carbon table.