Optimize Your Databases on Kubernetes With This One Weird Filesystem Trick

When: Thursday, October 17th | 12pm - 1pm EST

Schedule: 1-Hour Free Webinar

Where: Online - Zoom Meeting


What if I told you that you can reduce the size of your databases by 50% and significantly increase their performance, without changing your hardware, without touching your database engine, queries, or indexes, in a completely transparent manner?

The “secret sauce” is ZFS and the OpenEBS LocalPV-ZFS CSI driver. There is a lot to unpack here!

ZFS is an advanced filesystem with a lot of sophisticated features. The one that is relevant here is transparent compression, which is particularly efficient on most database workloads, as it reduce the amount of data that needs to be read from and written to disk. This reduces in lower disk usage and higher throughput.

To leverage ZFS transparently on Kubernetes, we will use LocalPV-ZFS. This is a CSI driver that allows automatic provisioning of Kubernetes volumes on a ZFS pool. We will see how to install this on a Kubernetes cluster.

Finally, to demonstrate and benchmark the whole setup, we will use CNPG - Cloud Native PostGres, a database operator that can provision replicated PostgreSQL database clusters in mere minutes.

During the Webinar:

Our goal is to demonstrate how multiple high-quality open source solutions can interoperate together to realize significant cost savings and performance improvements.

That said, each of these components is already useful and meaningful in isolation; so feel free to join the presentation even if you’re interested by only one of the components: ZFS, OpenEBS, or CNPG, for instance.

There will be a lot of live demos!


About Jérôme Petazzoni:

Jérôme was part of the team that built, scaled, and operated the dotCloud PAAS before it became Docker. He worked for seven years at the famous container company, wearing various hats. When he’s not busy with computers, he collects musical instruments. He can arguably play the theme of Zelda on a dozen of them.