AIOps Is How We Will Survive DevOps
summary
Claims that AIOps reduces 2am firefights by automating baselining, clustering and root‑cause triage, demonstrates a linear‑regression demo in pandas and Jupyter with a 70/30 split for train/test, and finds finished square footage surprisingly the strongest correlate in a San Francisco home‑price set. Shows AppDynamics‑style use cases: automatic thresholds from historical and contextual metrics, cluster‑based triage, and five concrete automation criteria—frequency, impact, coverage, probability, timescale—so the talk emphasizes practical tools, numbers, and repeatable steps rather than vague hype.
AI-written summary · about
chapters (8)
- 0:00Start
- 4:27Chapters
- 10:43Basic Terminology
- 16:12What is Normal? Goodbye Manual Thresholds!
- 18:47AlOps: Automated Root Cause(s) Analysis
- 21:46Frequency
- 23:00Coverage
- 27:59Questions & Links
related talks
19mAIOps
33mAge of AI Ops
42mAIOps for Distributed Environments - Deep Dive
20mAutomated Incident Management
30mIntelligent Analysis on Cloud Infrastructure With AIOps
33mAIOps-driven cloud automation with Dynatrace observability platform
33mAIOps-driven cloud automation with Dynatrace observability platform
02mAIOPS – How can machine learning help in IT operations