Green AI with JobRunr: Sustainable Training and Inference on the JVM
summary
Demonstrates JobRunr carbon-aware scheduling inside a Spring Boot app that collects session feedback, bundles comments, and enqueues a summarization job for a local LLM; demo uses Ollama + LangChain local LLM with GPT OSS 20B, shows the dashboard choosing a 3:00 low-carbon window and the job completing successfully, and explains using ENTSO-E forecasts and dynamic pricing to run non‑urgent training/inference when grid intensity is lowest, cutting CO₂ and cloud cost in practice.
AI-written summary · about
chapters (7)
- 0:00Introduction to Green AI
- 3:51Why Green AI matters
- 6:40Carbon-aware scheduling with JobRunr
- 15:26Live demo of feedback summarization
- 18:09Audience feedback collection
- 19:18JobRunner dashboard walkthrough
- 21:53Q&A on green energy scheduling
related talks
17mGreen AI with JobRunr: Sustainable training and inference on the JVM
29mHow AI and Deep Learning could effort the Climate Change
17mGREEN: Carbon-efficient Resource Scheduling for Machine Learning Clusters
07mLightning Talk: Environmentally Sustainable AI via Power-Aware Batch Scheduling
32mOptimizing AI/ML workloads for sustainability
50mProduction-Ready GenAI with Open Models for Java Teams
20mBuilding GenAI That Doesn't Cost the Earth
01h 02mJoint Keynote Address: Scaling AI Sustainably: An Uncharted Territory