Skip to content

Red Hat OpenShift Architecture Workshop

IBM Academy of Technology Red Hat OpenShift solution design guidance provides step-by-step guidance to conduct an OpenShift Decision Workshop, including:

The following diagram shows a sample on premises deployment reference architecture.

on premises Openshift architecture

The common reference architecture is a great place to start. But clients will want a deeper understanding of the deployment and its capabilities.

Red Hat Openshift deployment architecture

Here is a typical design pattern for Red Hat OpenShift solutions. In this pattern, the entire cluster (i.e., control plane, worker, infrastructure) exist in the same network zone.

The following conceptual deployment architecture showing three types of workloads:

  • containerized workloads from IBM Cloud Paks deployed on the Red Hat OpenShift cluster
  • containerized workloads from customer application microservices
  • the workload from non-containerized enterprise applications hosted on regular VMs.

Deployment architecture

The following operational model shows the physical view for a common deployment.

Physical architecture

Architecture decision point guidance

Cloud Solution Architects need to make these critical decisions as part of every Red Hat® OpenShift® design.

  • HA and DR
  • Red Hat OpenShift Cluster Configuration
  • Red Hat OpenShift Instance Size
  • Oversubscription

Architecture decision points can help guide you through the decisions in deploying Openshift. Decisions rationale, motives, and considerations are explained for each decision:

AD-001 - Red Hat OpenShift Cluster Platform Node Type
AD-002 - Hosting Platform
AD-003 - Red Hat OpenShift Platform Selection
AD-004 - Red Hat OpenShift Version
AD-005 - Master Node Scheduling
AD-006 - Cluster Workload Isolation
AD-007 - Master Node High-Availability
AD-008 - Application Availability
AD-009 - Load Balancer Selection
AD-010 - Multi-Region Red Hat OpenShift Cluster Deployment
AD-011 - Container Registry
AD-012 - Management Service Placement
AD-013 - Worker Node OS
AD-014 - Overcommitting Nodes
AD-015 - Editing Kubelet Parameters
AD-016 - Storage Technology for Red Hat OpenShift Cluster Platform Registry
AD-017 - Storage Technology for Metrics & Logging
AD-018 - Persistent Storage Options for Applications
AD-019 - Production and Non-Production Co-Location
AD-020 - Number of Installer Nodes
AD-021 - Ingress Traffic for Applications
AD-022 - High availability of UAT Environment

Reference Cluster Configuration

When you install OpenShift, you may want to include Operators and funcationality, such as:

  • OpenShift Container Storage
  • Cluster Logging
  • Git (using the Gitea operator)
  • Jenkins
  • OpenShift Pipelines
  • OpenShift GitOps (for clusters at 4.7+)
  • Quay
  • Nexus Repository (Optional - Artifactory is another common choice)
  • Red Hat Single Sign-on (Optional)
  • Red Hat Service Mesh (Optional)