Next Generation of DevOps AIOps in Practice
summary
Baidu builds an AIOps platform that standardizes operations across thousands of apps and hundreds of thousands of servers, claiming a unified operations language, an OKB (operations knowledge base) and an OPAL operations abstraction layer that expose APIs for reusable tools. It runs anomaly detection and traffic‑shifting algorithms to contain failures to a single cluster, adds an automatic checker for unattended deployments that improves success rates, and uses NLP to turn questions into structured operational queries.
AI-written summary · about
chapters (18)
- 0:00Intro
- 1:51Agenda
- 4:00Team
- 6:54Core Components
- 8:05Framework
- 9:44Operation Knowledge Base
- 10:33Data Mining Machine Learning
- 11:37PrePractice
- 12:45Failures
- 17:36Capacity redundancy
- 18:02Traffic shifting
- 19:15Access resistant strategy
- 20:11Single classes of loss
- 24:26Anomaly detection
- 25:33Single cluster AIOp
- 28:16Automatic deployment AIOp
- 30:07Natural language processing
- 32:04Summary
related talks
29mIs AIOps the Future of Operations? Real Use Cases From the Trenches
34mWhat If the Promise of AIOps Was True?
02mAIOPS – How can machine learning help in IT operations
24mAI Ops and its challenges
40mAI-Augmented DevOps with Platform Engineering
28mAIOPS the power of DevOps at scale
17mOptimizing Multi-Cloud Platforms with AI Ops and FinOps
17mAI TechOps: Data and Infra for AI Foundation to Scale and Manage 1000s of Cohesive Agents