Toward a carbon-aware Cloud
summary
Cloud infrastructure emits over 300 megatons CO2 today—roughly a country like France or the UK—and region choice can cut emissions by 10–20x (demo moves a Terraform-defined fleet from Germany to Paris, dropping ~600 kg/yr to ~100 kg/yr). Measurement combines per-process power probes, PUE, embodied emissions and grid carbon intensity from ElectricityMaps, billing readers like Cloud Carbon Footprint and APIs such as climatic or bav Vista. A live carbon-aware scaler demo shows shutting down greener regions and a case study reports ~21% reduction with acceptable latency.
AI-written summary · about
chapters (9)
- 0:00Carbon-aware cloud intro
- 2:34Why cloud carbon matters now
- 10:47Regulations and reporting
- 18:04Measuring carbon footprint with SCI
- 24:12Tools for emissions reduction
- 31:17Green coding and optimization
- 39:07Carbon-aware scaling strategies
- 43:22Demo: carbon-aware load balancing
- 50:35Key takeaways and next steps
related talks
41mToward a carbon-aware cloud
51mTowards a Carbon-Aware Cloud
41mWriting Greener Software Even When You Are Stuck On-Prem
42mUndertaking a GreenOps Approach in the Public Cloud
34mEmbracing Sustainability in Cloud-Native Ecosystems
32mGreenOps in the cloud
28mHow to limit your cloud impact, from bare metal to AI
01h 00mSustainable Azure - Tools & Patterns for Carbon Aware Cloud Applications