How to Cloud Ops Like a Boss
summary
Panelists argue enterprises must stop treating Cloud Foundry platforms like legacy deployments and adopt concrete patterns: hand deploys to developers, CI/CD pipelines, blue‑green deploys, and platform continuous restore to automate rollbacks. They show tools—PCF, Dynatrace, Sumo Logic—used for automated root‑cause analysis and log anomaly detection with machine learning, recommend per‑app smoke tests and API‑to‑API SLOs, note ~12 expert troubleshooters bottleneck scaling from ~20 to 200 apps, and aim to cut MTTR.
AI-written summary · about
chapters (10)
- 0:00Introduction to Cloud Ops
- 1:06Panelist introductions
- 2:31Enterprise anti-patterns in cloud ops
- 7:51Cultural changes for cloud-native adoption
- 12:08Best practices for cloud operations
- 17:44Tool sprawl and unified monitoring
- 23:29Measuring ops success beyond MTTR
- 28:07Biggest unsolved cloud ops problems
- 31:42Improving operator community communication
- 33:53Closing remarks and resources
related talks
24mCracks in the Foundation: Operating Cloud Foundry in the Real World
50mCracks in the Foundation: Operating Cloud Foundry...
32mOps in the Time of Serverless Containerized Webscale
43mHow Platforms Work
24mHow to survive Cloud – An Ops perspective
16mDeployments We Can Believe In
49mAutomation in Deployment on Hybrid Hosting and Private Cloud Environments
19mKeynote: Confessions of a Cloud Foundry PaaS Product Owner: How to Create a Successful Devops Team