Are Containers Ready for Production Databases?
summary
Argues that Docker/Podman-style containers can run production databases but success depends on architecture and scale, recommends Kubernetes when teams use operators like Percona’s for backups, upgrades, failover and accept extra day‑two work. Cites a community survey: ~90% say Kubernetes is ready for stateful workloads and 71% run them in production. Offers a practical size rule: treat databases >1 TB cautiously and avoid moving monolithic >4 TB systems into Kubernetes without re‑architecting.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 1:50Containers and databases hot topic
- 3:35Linux containers history
- 7:30Containers and Docker tools
- 9:37Kubernetes adoption and stateful workloads
- 12:28Databases on Kubernetes pros and cons
- 15:13Day two operations challenges
- 18:03Kubernetes operators for databases
- 21:12When Kubernetes is not the right choice
- 23:38Q&A
related talks
45mBest Practices for Running Databases on Kubernetes
36mFrom a Database in Container to DBaaS on Kubernetes
47mFrom a Database in Container to DBaaS on Kubernetes
45mDatabases on Kubernetes: Why you should care
38mThe Container Operator’s Manual
01h 06mPanel: Running Databases in Containers
34mDatabases on Kubernetes
26mChallenges of Managing Stateful Applications in Kubernetes