Strategic Capacity Planning & Forecasting for OpenShift at Scale

Welcome to the Workshop

Welcome to the Strategic Capacity Planning & Forecasting Workshop. This full-day, highly interactive session is designed to transition platform engineering and development teams from reactive "firefighting" to predictive, data-driven capacity management.

Using OpenShift and Red Hat Advanced Cluster Management (RHACM), you will explore the financial impacts of node density, resolve the tension between developer resource requests and infrastructure safety buffers, and build centralized, multi-cluster capacity dashboards.

Who Will Benefit Most?

Platform Engineers & SREs

Learn fleet-wide forecasting, node density optimization, and how to build custom RHACM observability dashboards.

Application Developers & Architects

Understand the Kubernetes scheduler, QoS classes, and why "accurate requests" are essential for stability rather than just administrative hurdles.

FinOps & Product Owners

Gain visibility into cloud spend, chargeback modeling, and the financial impact of commitment strategies.

Prerequisites

Before beginning this workshop, you should have:

  • Basic familiarity with the OpenShift/Kubernetes CLI (oc or kubectl)

  • Basic understanding of Prometheus (PromQL) and Grafana

  • Access to the Workshop Lab Environment (already provided)

Workshop Environment

Single-User Workshop Mode

You are using a shared workshop environment on the hub cluster.

Your Workshop Assignment:

  • Namespace: capacity-workshop

  • Workshop GUID: workshop-guid

What This Means:

  • You have namespace-scoped permissions

  • All exercises run within your dedicated namespace

  • Some infrastructure labs may have limited permissions (Module 4)

Your lab environment includes:

  • OpenShift Cluster: apps.example.openshiftapps.com

  • Namespace: capacity-workshop

  • GUID: workshop-guid

Quick Access Links

Component URL

OpenShift Console

https://console-openshift-console.apps.example.openshiftapps.com

RHACM Console

https://multicloud-console.apps.example.openshiftapps.com

Grafana Observability

https://grafana-open-cluster-management-observability.apps.example.openshiftapps.com

Connecting to Your Student Cluster

This workshop uses a hub-student topology: the lab guide (Showroom) runs on a central hub cluster, while your oc commands run on your own dedicated student cluster. The terminal on the right connects you to the hub — to reach your student cluster, SSH to its bastion host.

Step 1 — SSH to your student cluster bastion

Run the following command in the terminal on the right to connect to your student cluster:

ssh lab-user@

SSH password:

Once connected, verify your cluster access:

# Confirm you are logged into the student cluster
oc whoami --show-server

# Check cluster nodes (expect 3 compact nodes)
oc get nodes

# Verify workshop namespace and sample apps
oc get pods -n capacity-workshop
Hub-Student Topology: What Runs Where

  • This terminal — runs on the hub cluster (Showroom). Use it to SSH to your student bastion and for browser links.

  • Your student cluster — — runs all workshop sample apps, Prometheus, and simulation tools. All oc commands throughout the modules target this cluster.

  • RHACM + Grafana — run on the hub cluster and provide a fleet-wide view across all student clusters. You access them via browser (Quick Access Links above) during Module 5.

You should see after connecting:

  • 3 nodes when running oc get nodes (compact 3-node cluster)

  • Cluster-admin access — you can view all namespaces with oc get projects

  • Sample applications running in the capacity-workshop namespace

Workshop Agenda

Time Section Description Type

09:00

Module 1: The Planning Horizon

Tactical vs. Strategic planning; assessing current state

Lecture/Lab

10:00

Module 2: The Math of Forecasting

Pod Velocity, PromQL, and ACM Forecasting Dashboards

Lecture/Lab

11:00

Break

11:15

Module 3: Developer Track

Sizing, QoS, and the danger of "Zero Requests"

Lab

12:45

Lunch

13:45

Module 4: Infra Track

Multi-cluster sizing, density, and etcd limits

Lab

15:15

Module 5: Fleet Observability

Custom multi-cluster Grafana capacity dashboards

Lab

16:15

Break

16:30

Module 6: The Integration Challenge

Live simulation: Product launch spike + AZ failure

Game/Lab

17:30

Module 7: Strategic Roadmapping

Building the 12-month capacity plan

Exercise

18:00

Wrap-up & Q&A

18:30 (optional)

Module 8: AI-Assisted Capacity Operations (Optional)

OpenShift Lightspeed as a forecasting co-pilot — developer, infra & planning queries; granite vs. qwen3 model comparison

Lab

Technology Stack

This workshop uses:

  • OpenShift Container Platform: 4.21+

  • Red Hat Advanced Cluster Management (RHACM): 2.16+

  • Monitoring: Prometheus, Thanos, Alertmanager, Grafana

  • AI Assistant (Module 8, optional): OpenShift Lightspeed 1.0 backed by RHDP LiteMaaS (IBM Granite 3.2 8B / Qwen3 14B)

Getting Started

SSH to your student cluster bastion (see Connecting to Your Student Cluster above), then proceed to Module 1: The Planning Horizon to begin your capacity planning journey.

Before moving to Module 1, make sure you have an active SSH session to your student bastion: ssh lab-user@
All oc commands throughout the workshop run inside this SSH session.