Operate First Environment#

drawing

Storage#

Some network-attached storage. E.g. a Ceph cluster. For experimental purposes, the storage serving layer can change (ceph -> rook, gluster, nfs)

We do not run a registry, but assume that quay.io is used for all released container images. Each OpenShift cluster will have its own persistent internal registry.

Orchestration / Management Cluster#

Use case#

Run supporting tools to deploy, monitor and orchestrate the various environments. E.g. ArgoCD, external Monitoring, Ticket Management, etc.

Hardware#

14 VCPU, 64GB RAM, 1 TB storage

Dev#

Use case#

Spin up experimental minimal OCP clusters to develop in an interactive environment.

Hardware#

[4 VCPU, 32GB RAM, 50 GB storage] / Dev Cluster

OSP Platform to provide this HW as virtual resources

Test & Build (CI/CD/CD)#

Use case#

Run automated unit and integration tests. Build various assets, general tekton pipelines

Hardware#

64 VCPU, 512 GB RAM, 1.5 TB storage

Stage N#

Use case#

Minimal replica of production environment with running test workloads. N implementations, depending on different platform configurations and different workloads.

Hardware#

[16 VCPU, 128GB RAM, 500 GB storage] / Stage Cluster

Production#

Use case#

Long-running stable environment to support end-user workloads.

Hardware#

128 VCPU, 2 TB RAM, 400 TB storage