Running PostgreSQL on Kubernetes: Is It a Blessing or a Curse?
Thursday, June 25 at 13:20–14:10
-
Independent PostgreSQL consultant currently working with the European Patent Office (EPO). Holds a B.S. in Computer Science and has over 20 years of experience in database systems.
Containers and Kubernetes are now common for running applications, but many database teams are still cautious about running PostgreSQL on Kubernetes. For most DBAs, virtual machines are a well-understood and predictable environment, while Kubernetes introduces an operational model that is not always obvious at first.
In this session, we look at what actually changes for a DBA when PostgreSQL runs on Kubernetes. Based on real production experience, the talk compares everyday tasks such as backups, upgrades, monitoring, and failover with more traditional PostgreSQL deployments on virtual servers. The aim is to share where Kubernetes can help simplify operations and where it introduces new responsibilities and trade-offs.
Target Audience: DBAs, architects, DevOps engineers, and technical decision-makers
Technical Level: Mix of low-level technical details and high-level architectural concepts
Talk Outline:
- From bare metal servers to containerization: evolution of virtualization technologies
- Fitting the elephant inside a container: how containerized PostgreSQL differs from traditional deployments
- Running PostgreSQL in production on Kubernetes: Core production concerns: high availability, monitoring, security, backup and recovery
- Has the time come to leave conventional virtual servers? Challenges of administering PostgreSQL on Kubernetes: what to expect in real-world projects. How Kubernetes can ease a DBA’s life: PostgreSQL operators
