WEKA CSI Plugin

Connect Kubernetes worker nodes to the WEKA data platform to leverage its capabilities.

The WEKA CSI Plugin interfaces Kubernetes worker nodes and control plane with the WEKA clients. It enables the creation and configuration of persistent storage external to Kubernetes. It is based on the CSI Plugin standard.

What is the CSI Plugin standard?

The CSI (Container Storage Interface) Plugin is a standardized interface that enables container orchestration platforms, such as Kubernetes, to interact with different storage systems in a vendor-agnostic manner. CSI was introduced to address the challenges of integrating and managing storage in containerized environments.

For more details, see the CSI standard specifications.

The WEKA CSI Plugin provides the following features:

  • Static and dynamic volumes provisioning.

  • Mount a volume as a WEKA filesystem directory, a writable snapshot of an existing filesystem, or a whole filesystem.

  • Support all volume access modes: ReadWriteMany, ReadWriteOnce, and ReadOnlyMany.

  • Volume expansion.

  • Snapshots and volume cloning.

  • Quota enforcement on:

Interoperability

  • CSI protocol: 1.0 - 1.2.

  • Kubernetes: 1.18 - 1.3x.

  • WEKA: V3.14 and up. WEKA V4.2 and up and WEKA CSI Plugin V2.0 are required to get all features.

  • SELinux is supported (AppArmor for Ubuntu is not supported).

  • Starting with WEKA CSI Plugin version 2.5.0, NFS transport is available as a lower-performing alternative to WekaFS. This option is intended for non-performance-critical scenarios or environments where installing the WEKA client is not feasible.

The WEKA CSI Plugin replaces plugins developed earlier in the Kubernetes evolution. It replaces the hostPath method to expose WEKA mounts as Kubernetes volumes.

Advanced container deployment using the WEKA CSI 2.0 Plugin

This demo showcases the process of deploying advanced containerized applications using the WEKA CSI 2.0 Plugin. It highlights key features and configurations that can optimize your container environment.

Last updated