Sustainable Software Engineering & SREs
summary
Planetary SLO sits at 1.5°C and current warming is about 1.15°C with roughly 12 years before breach; software choices affect that. Examples include cutting cores and improving scaling, better indexing and sharding to save power, co‑locating dependent microservices, delaying batch work to low‑carbon windows using WattTime signals, and adding carbon‑aware scheduling in Kubernetes. A 213M‑parameter NLP run costs ~620,155 lb CO2; track models with mlco2impact and follow principles.green guidance. Set measurable SLOs, log carbon intensity, and prioritize efficiency.
AI-written summary · about
chapters (10)
- 0:00Introduction to sustainability
- 1:24Defining the planet's SLO
- 2:36SRE and sustainability
- 4:20Balancing system trade-offs
- 5:58Core principles of efficiency
- 7:49Practical implementation
- 11:08The impact of AI
- 13:42Efficiency and the environment
- 15:45Corporate climate progress
- 17:03Call to action
related talks
23mSLO's for Openstack - the API for Engineering Teams
35mCare and Feeding of SRE
25mPitching SRE to your Senior Executives
05mSLOs: You’re missing the point!
25mSRE Guiding Principles & Transformational Framework
07mIGNITE: Metrics, Margins, Mutiny – How to make your SREs (not) run away
41mAn Introduction to SRE at Google (with DevOps Focus)
40mBuilding Successful SRE in Large Enterprises—One Year Later