Presented by:

Fernando has been working with Linux and open source for the past 20 years, first in a DevOps role and later specializing in database technologies. He joined Percona in 2013 and has been focusing on the universe of PostgreSQL and MySQL with a pinch of Kubernetes.

No video of the event yet, sorry!

"Nothing", you might be thinking. At least I used to: it's okay for deploying application servers in an "elastic" capacity, but not for databases, which are meant to be stable and faultless. However, while Kubernetes environments remain very dynamic, it evolved to support stateful applications better and now counts with an enthusiastic Data on Kubernetes (DoK) community that keeps increasing in size. Why is that so? Much like DBaaS, I think ease of deployment and management can be a hefty trait of Kubernetes - that is, when one uses a specific operator for the job!

In this talk, I will show you how to deploy a complete PostgreSQL environment on Kubernetes packed with some of your favorite tools: Patroni for high availability, pgBackRest for backups and WAL archiving, and pgBouncer for connection pooling. We add in monitoring with PMM for good measure and run some benchmarks to stress test the new environment.

This may look like a how-to disguised as a discussion on the topic, and it is in a way: even if you are not familiar with Kubernetes and this is your first time looking at it, we can get you up and running with a short procedure. But the main goal is to use it to present what I find attractive in PostgreSQL on Kubernetes as well as highlight what are the main challenges you should expect to face. Then you can decide it's really not for you; or maybe it could be?

Date:
2024 November 6 12:10 PST
Duration:
50 min
Room:
Ops: 421
Conference:
Seattle 2024
Language:
Track:
Ops
Difficulty:
Medium