Key numbers
gCO2e per pipeline run
App scope (lint, tests, image build). Baseline = the intentionally wasteful pipeline; optimized = after the Duo optimizer flow's MR merged.
No pipeline measurements in the ledger yet.
Latest deploy decisions
| When (UTC) | Decision | Why | Grid |
|---|---|---|---|
| No decisions yet. | |||
Projection projection, not a measurement
If a team ran this optimization on , the measured saving per pipeline would add up to:
–
How this is measured
SCI = ((E × I) + M) per R, with R = one pipeline run (ISO/IEC 21031:2024). E comes from job runtime × vCPU and memory power coefficients (Cloud Carbon Footprint) × PUE, I is the grid intensity of the runner region (Google Cloud us-east1, where GitLab.com hosted runners run), M is amortized embodied hardware carbon. Absolute grams are estimates; the percentage reduction compares the same jobs with the same method.