The role of data in green software engineering
summary
Data already consumes about 200 terawatt hours a year — more CO2 than the entire airline industry — so the talk gives actionable rules: eliminate utilize allocate. Specifics include dropping unnecessary backups, removing roughly 30% duplicate data found in one audit, avoiding overgenerated image variants, using cloud provider carbon dashboards and FinOps to rightsize Kubernetes, preferring greener regions, and calling the Green Software Foundation API to schedule work. These steps cut both energy and cost, a clear win when implemented.
AI-written summary · about
chapters (14)
- 0:00Data as the new oil
- 1:44Energy impact of data centers
- 3:55Daily engineering challenges
- 4:59Three rules for sustainable data
- 8:54Pareto principle for data
- 11:02Strategic compute allocation
- 13:11Energy-efficient cloud regions
- 15:17Education in green software
- 16:02Sustainable engineering targets
- 18:32Sustainable tech across layers
- 19:15New business models
- 20:37Programming languages for efficiency
- 25:56Green software resources
- 29:29Summary and call to action
related talks
46mMaking Software & Data Architectures Sustainable
41mWriting Greener Software Even When You Are Stuck On-Prem
57mSustainable Green Software development
38mGreen software: How can tech contribute to worldwide sustainability effort?
43mGreen Software – How To Make Our Applications Carbon Aware?
55mCode that Won't Heat the Planet: Engineering Sustainable Software for a Better Future
28mSustainable IT : Connect the Dots to a Greener Future
24mGreen IT is good but it's not enough!