Technical Research Fueling NextGen Actions
summary
Research shows data centers’ space-time flexibility can cut renewable curtailment and deliver very low-emission consumption portfolios with far fewer resources than expected, while power flow analyses using power flow tracing on transmission models find most electricity is consumed within ~100 km though exports (Norway/Sweden) can travel much farther. Counterintuitively, some common load‑shaping heuristics increase annual CO2. Tools and outputs are open source (GitHub, pipes.org), models simulate ERCOT operations, and computing demand has grown >6x in 10 years and now uses ~1% of global electricity.
AI-written summary · about
chapters (12)
- 0:00Introduction and welcome
- 3:25Why research matters now
- 6:35Researchers' personal motivations
- 11:5624/7 clean energy procurement study
- 15:44Power flow tracing on grids
- 18:54Load shaping for decarbonization
- 23:05Synergy between research efforts
- 24:50Open source in energy research
- 26:58Advice for the audience
- 28:47What's next in research
- 32:20Closing and future studies
- 33:24Q&A session
related talks
20mBuilding GenAI That Doesn't Cost the Earth
37mEnergy Consumption of Datacenters
41mSustainable Digital
47mSoftware and Climate: build a better world
36mOnline Learning and Optimization in Distributed Energy Systems: Some Problems and Opportunities
27mOpen Source in an Age of Transformation for the Energy Industry
26mAn EV Boom or a Gridlock? A Data-Driven Approach to UAE's Green Energy Transition
35mSustainability Research the Cloud Native Way