Sustaining Everything, Everywhere, All at Once!
summary
IBM builds a full‑stack sustainability platform running in a 20‑year Beijing data center that collects environmental sensors and agentless telemetry from servers, storage and apps, correlates energy use to transactions, and issues automated actions (shut down, migrate, resize, suspend). A dynamic electricity‑aware scheduler shifts throughput to low‑price windows (example cluster caps at 96 jobs -> 48 jobs during high prices), adds predictive maintenance and AR remote repair, and exports metrics via Apache Kepler and Grafana — demonstrably improving PUE, IT energy and emissions.
AI-written summary · about
chapters (6)
- 0:00Introduction
- 5:08IBM Academy sustainability role
- 12:10Full-stack sustainable computing
- 24:23Live demo: sustainability dashboard
- 36:47Open-source plans
- 42:53Q&A
related talks
35mSustainable Platform Engineering: Having a Green Thumb Without Shooting Up
26mAccelerating Science, Decelerating Carbon: Sustainable Computing...
26mWhy sustainability makes sense for tech
42mBuilding a green supercomputer in the cloud
41mWriting Greener Software Even When You Are Stuck On-Prem
06mTalking Open Source & Cloud Native
57mOptimizing your infrastructure for sustainability- 26mWhy sustainability makes sense for tech