Managed Storage Cluster
The Managed Storage Cluster offers advanced cloud-native storage capabilities for Managed OpenShift.
It brings a huge list of features, including ReadWriteMany
storage volumes and is meant as an add-on where the capabilities of the CSI storage integration of the cloud provider isn’t sufficient to meet the needs or for infrastructures where there is no CSI integration possible.
Supported Features and Configuration
Supported by default
These features and configurations are available out-of-the box and installed and configured by default.
Feature | Description |
---|---|
Managed Rook |
The Rook storage operator is installed, configured and maintained. |
|
These access modes are available via Ceph. |
|
This access mode is available via CephFS. |
In-Cluster Object Storage |
With the Rados Gateway of Ceph an in-cluster S3-compatible object storage is available. |
Dynamic Provisioning |
Volumes are dynamically provisioned and request by creating a PersistentVolumeClaim. |
Metrics and Monitoring |
All key metrics are collected and monitored on if there are any anomalies. |
Volume Snapshots |
Kubernetes Volume Snapshots are supported. |
Volume Cloning |
Cloning of volumes is fully supported. |
Volume Resizing |
Dynamic volume expanding is available. |
Replication and High-Availability |
The default replication factor is 3 (meaning that 3 copies of any given data is stored in the cluster) and if possible the storage cluster nodes are distributed over different availability-zones and are configured using anti-affinity virtual machine rules. |
Storage Encryption at Rest |
By default the Rook Ceph cluster is configured with |
Supported on request
These features or configuration adjustments must be specifically requested and some restrictions apply. Activation and configuration of these features imply additional engineering costs and can cause additional engineering costs for operating them (although no fixed additional recurring costs apply).
Feature | Description |
---|---|
In-Cluster Object Storage |
With the Rados Gateway of Ceph an in-cluster S3-compatible object storage is available. |
Unsupported
These features or configuration adjustments are not supported by VSHN, but can still be activated or changed, allthough are neither monitored, backed up nor maintained. No guarantees are given, use them at your own risk.
Feature / Configuration | Description | Reasoning |
---|---|---|
Encryption per PV |
Per volume encryption as supported by the Ceph CSI driver. |
This feature is very complex in operations and doesn’t automatically bring added value over the VSHN supported storage encryption at rest as it uses the same encryption mechanisms. |
Still interested in one (or more) of these unsupported options? Get in contact with sales@vshn.ch and we figure out together what we can offer. |
Pricing
It’s included in the Managed OpenShift fee.
The vCPUs of the storage nodes count as "Worker-vCPU", but these nodes can only run the storage cluster components, no other workload can run on these nodes.
Default Configuration
By default, 3 storage nodes will be provisioned. This is the minimum supported configuration, it’s not possible to have less than three storage nodes.
The sizing of the storage nodes depends on the infrastructure.
Limits and Scaling
- Max. number of Persistent Volumes
-
With 3 storage nodes, max. 200 Persistent Volumes are supported. More volumes are available on request and require additional storage nodes.
- Storage Node Sizing
-
A storage node needs at least:
-
8 vCPUs
-
16 GB RAM (24 GB RAM if you’re planning to use CephFS)
-
120 GB storage for the OS
-
512 GB storage for the storage cluster
-
- Rook Storage Provider
-
Only the Ceph storage provider is supported. Other Rook storage providers aren’t supported (e.g. Cassandra or NFS).