Making Software & Data Architectures Sustainable
summary
Claim: cloud computing is about 3% of global greenhouse gases (roughly 1.5× aviation and 3× UK annual emissions) and migrating appropriate workloads can cut ~40 megatons CO2e. Practical levers include the Green Software Foundation and AWS sustainability pillar, tooling like CodeCarbon, Green Algorithms and Element AI estimators, and hardware choices (TPU vs GPU). Concrete actions: delete unused Kafka topics, stop continuous retrains, collocate coupled services, trim payloads (saving 1 MB per order on 150M orders can cut ~4 metric tons/year).
AI-written summary · about
chapters (11)
- 0:00Introduction and context
- 2:46Evolution of computing
- 6:48Modern IT and climate impact
- 12:00The data challenge
- 15:49Shared responsibility
- 19:28Carbon intensity equation
- 21:54Optimizing computation
- 25:19AI energy consumption
- 31:27Communication and transport
- 35:33Carbon-aware infrastructure
- 38:55Actionable strategies
related talks
55mCode that Won't Heat the Planet: Engineering Sustainable Software for a Better Future
57mSustainable Green Software development
18mEngineering for Sustainability
41mWriting Greener Software Even When You Are Stuck On-Prem
30mThe role of data in green software engineering
44mCloud computing and carbon emissions
38mGreen software: How can tech contribute to worldwide sustainability effort?
28mSustainable IT : Connect the Dots to a Greener Future