Cloud Integration Guide
Connect your cloud infrastructure to automatically track compute, storage, and network emissions across AWS, Azure, GCP, and more.
What You'll Learn
- Cloud provider setup and permissions
- Service account configuration
- Billing integration and cost allocation
- Emission factor configuration
- Regional data center mapping
- Monitoring and optimization
Prerequisites
Before you start, ensure you have:
Account Requirements
- Active Carbon Recycling Platform account
- Cloud provider administrator access
- Billing account access (for cost data)
Information Needed
- Cloud account IDs and subscription details
- Primary regions and data centers used
- Resource tagging strategy (if any)
Choose Your Cloud Provider
We support 20+ cloud providers from major hyperscalers to specialized platforms.
Amazon Web Services
- • EC2, S3, RDS, Lambda
- • Cost & Usage Reports
- • CloudWatch metrics
- • Multi-account support
Microsoft Azure
- • VMs, Storage, SQL, Functions
- • Cost Management API
- • Azure Monitor
- • Subscription hierarchy
Google Cloud Platform
- • Compute Engine, Cloud Storage
- • Cloud Billing API
- • Cloud Monitoring
- • Project organization
Additional Cloud Providers
DigitalOcean
Droplets, Spaces, App Platform
Cloudflare
Workers, R2, Pages
Vercel
Serverless functions, Edge
Netlify
Functions, CDN, Forms
Multi-cloud setup? You can integrate multiple providers simultaneously. We'll help you consolidate emissions data across all your cloud infrastructure.
Set Up Service Account Authentication
Amazon Web Services Setup
Create IAM Role for Cross-Account Access
- Log into your AWS Management Console
- Navigate to IAM → Roles → Create Role
- Select "Another AWS Account"
- Enter Account ID:
123456789012(our Carbon Recycling account) - Require external ID:
[provided during setup] - Attach the following policies:
- ReadOnlyAccess
- CostExplorerServiceReadOnlyAccess
- CloudWatchReadOnlyAccess
- BillingReadOnlyAccess Enable Cost and Usage Reports
- Go to AWS Billing → Cost & Usage Reports
- Click "Create Report"
- Report name:
carbon-recycling-cur - Include resource IDs: Yes
- S3 bucket: Create new or use existing
- File format: Parquet (recommended)
Microsoft Azure Setup
Create Service Principal
- Open Azure Cloud Shell or Azure CLI
- Run the following command:
az ad sp create-for-rbac --name "carbon-recycling-reader" \
--role "Cost Management Reader" \
--scopes "/subscriptions/YOUR-SUBSCRIPTION-ID" Save the output JSON - you'll need the appId, password, and tenant values.
Assign Additional Permissions
- • Reader role on subscription level
- • Cost Management Reader for billing data
- • Monitoring Reader for metrics
Google Cloud Platform Setup
Create Service Account
- Go to IAM & Admin → Service Accounts
- Click "Create Service Account"
- Name:
carbon-recycling-reader - Grant the following roles:
- Cloud Asset Viewer
- Billing Account Viewer
- Monitoring Viewer
- Compute Viewer Generate Key File
- Click on the created service account
- Go to "Keys" tab
- Click "Add Key" → "Create new key"
- Choose JSON format
- Download and securely store the key file
Configure Data Collection & Mapping
Map your cloud resources to our carbon accounting framework for accurate emissions tracking.
| Service Category | Resources Tracked | Emission Scope | Calculation Method |
|---|---|---|---|
| Compute | VMs, Containers, Serverless functions | Scope 2 - Purchased electricity | Instance hours × region factor |
| Storage | Object storage, Block storage, Databases | Scope 2 - Purchased electricity | GB stored × storage factor |
| Network | Data transfer, CDN, Load balancers | Scope 3 - Downstream transport | GB transferred × network factor |
| Managed Services | AI/ML, Analytics, Monitoring | Scope 3 - Purchased services | Usage metrics × service factor |
Regional Emission Factors (kgCO2e per compute hour)
Test Your Integration
Validation Checklist
Authentication & Permissions
- Service account connects successfully
- Billing data is accessible
- Resource metrics are retrievable
- Multi-region access works
Data Quality & Accuracy
- Resource usage aligns with console
- Cost data matches billing reports
- Emission calculations are reasonable
- Historical data imports correctly
Sample Test Results
Cloud Integration Test Results: AWS Production Account
✅ Authentication: SUCCESS (Role: arn:aws:iam::123456789012:role/CarbonRecyclingReader)
✅ Billing Access: SUCCESS (CUR reports available)
✅ Resource Discovery: SUCCESS (1,247 resources found)
✅ Regional Mapping: SUCCESS (3 regions: us-east-1, eu-west-1, ap-southeast-1)
Monthly Resource Summary:
- EC2 Compute Hours: 8,760 hours
- S3 Storage: 2.4 TB
- Data Transfer: 156 GB
- Estimated Emissions: 127.3 kgCO2e
- Monthly Cost: $1,847.23 Optimize & Monitor
Carbon Optimization Tips
- Choose regions with renewable energy grids
- Use auto-scaling to minimize idle resources
- Implement right-sizing recommendations
- Use serverless functions for variable workloads
- Set up lifecycle policies for storage
Monitoring Setup
- Configure alerts for emission spikes
- Set up monthly carbon reporting
- Track carbon efficiency metrics over time
- Review architecture recommendations
- Monitor for service updates and new features
Your Carbon Dashboard Will Show:
Common Issues & Solutions
Service account authentication fails
Possible causes:
- Incorrect permissions on the service account
- Cross-account trust policy misconfiguration
- Service account key file corrupted or expired
Solutions:
- Verify all required permissions are attached
- Check trust relationships and external ID (AWS)
- Generate new service account keys
- Test authentication using cloud CLI tools
Billing data not accessible
Possible causes:
- Missing billing permissions
- Cost & Usage Reports not enabled
- Billing account linked to different organization
Solutions:
- Add billing-specific roles to service account
- Enable detailed billing reports in cloud console
- Verify billing account access and permissions
High API usage or rate limiting
Possible causes:
- Too frequent sync intervals
- Large number of resources being queried
- Multiple integrations hitting same API
Solutions:
- Adjust sync frequency to optimal intervals
- Implement resource filtering and pagination
- Configure exponential backoff for retries
Emission calculations seem incorrect
Possible causes:
- Regional emission factors not updated
- Resource usage metrics misaligned
- Double counting across services
Solutions:
- Verify regional data center mapping
- Cross-reference usage with cloud console
- Review resource attribution settings
Need Help?
Our cloud specialists are here to help you set up and optimize your cloud carbon tracking.