GreenOps & Carbon Footprint

The GreenOps module provides a carbon footprint estimate for your cloud fleet. It is designed to help engineering and FinOps teams understand the environmental impact of their cloud infrastructure and identify opportunities to reduce emissions -- without requiring dedicated sustainability tooling.

Carbon estimates are derived from resource inventory and utilization signals. They are approximate -- not certified carbon accounting -- but are useful for prioritization, trend tracking, and surfacing high-impact resources.

Carbon is estimated using the following formula per resource:

Where:

• runtime_hours: number of hours the resource has been running (derived from scan data and resource age)
• cpu_utilization: average CPU % from the last available metric, or 50% default if unavailable
• region_emission_factor: kgCO2e per compute-hour for the resource's region

Region emission factors used (kgCO2e per compute-hour):

Factors are derived from public cloud provider sustainability reports and grid emission data. Per-region granularity is on the roadmap.

The GreenOps dashboard shows:

• Total estimated monthly CO2e across all scanned resources (in kgCO2e or tCO2e).
• Tree equivalent: how many trees would be needed to offset the annual estimated emissions.
• Car equivalent: annual emissions expressed as equivalent km driven.
• Provider breakdown: emissions per cloud provider.
• Region breakdown: which regions contribute most to your footprint.
• High-carbon resource table: individual resources with the highest estimated emissions.
• Green region suggestions: resources running in high-emission regions with available greener alternatives.

Practical steps to reduce cloud emissions using TurboFinOps data:

1. 1. Shut down unused resources

   The biggest emission reduction is eliminating resources that are running but not used. Use the Zombie Resources dashboard to find idle and orphaned resources and clean them up.

2. 2. Right-size over-provisioned instances

   Larger instances consume more power. Use the FinOps rightsizing recommendations to match resource size to actual workload needs.

3. 3. Move workloads to greener regions

   GreenOps surfaces resources in high-emission regions with lower-emission alternatives available. Migrating to eu-north-1, swedencentral, or europe-north1 can significantly reduce per-resource emissions.

4. 4. Use VM scheduling

   Stop compute instances outside business hours using the VM Scheduling feature. A resource that runs 8 hours/day instead of 24 produces ~66% less carbon.

• Carbon estimates are approximations based on available inventory data. They are not certified carbon accounting and should not be used for regulatory reporting without further verification.
• CPU utilization data is estimated as 50% when actual metrics are unavailable. Connect CloudWatch, Azure Monitor, or Cloud Monitoring for more accurate utilization signals.
• Emission factors are provider averages, not per-region real-time grid data. Per-region granularity with live grid data is on the roadmap.
• Storage and network transfer emissions are not included in the current calculation. Compute-based emissions are the primary component.
• Purchased renewable energy certificates (RECs) or carbon offsets held by cloud providers are not reflected in the current model.