Infrastructure requirements
To size an Azure deployment of OpenLM Platform, start from this baseline. Adjust these values to the scale and load of your environment.
Networking
Use Azure Container Networking Interface (CNI) mode for Azure Kubernetes Service (AKS). The following table lists the baseline network values:
| Item | Baseline |
|---|---|
| Network block | /22, such as 10.0.0.0/22 |
| Address capacity | 1,024 IP addresses |
| Planned running pods | About 143 |
| Planned nodes | 6 to 7 |
| Reserved addresses | About 400 IP addresses for pods, nodes, and growth |
Capacity example
The following points show how the network sizing works out:
- Each node allows about 29 pod IP addresses.
- Six to seven nodes provide about 174 to 203 pod IP addresses.
- A total of 143 pods and 6 nodes consume about 149 IP addresses.
- A
/22block leaves room for growth.
Azure Kubernetes Service
The baseline uses the following node pools:
| Node pool | Purpose | Node count | Max pods per node | Size |
|---|---|---|---|---|
| System node pool | OpenLM Platform system pods | 2 | 30 | 4 vCPU, 16 GB RAM, such as Standard_D4ds_v4 or a comparable v5 or v6 size |
| Main node pool | Core application workloads | 2 | 70 | 4 vCPU, 16 GB RAM |
| Reporting node pool | Reporting workloads | 3 | 30 | 4 vCPU, 16 GB RAM |
Use Kubernetes version 1.32.9 or another AKS-supported version that matches your release policy.
The baseline uses 7 virtual machines across the 3 node pools. Nodes with 16 GB of RAM leave headroom for OpenLM Platform pods because baseline node overhead is about 1 GB per node.
Managed services
The following table lists the managed Azure services and their baseline configurations:
| Service | Purpose | Baseline |
|---|---|---|
| Azure SQL Managed Instance | Relational database for identity, server, and reporting data | General Purpose, 4 vCores, 256 GB storage minimum |
| Azure Cache for Redis | Cache and session storage | C2 tier with 2.5 GB cache |
| Azure Managed Disks | Persistent volumes for Kafka, MongoDB, and other stateful services | About 400 GB total |
AKS provisions managed disks automatically when the cluster uses the required storage classes.
Sizing notes
Review the following sizing guidelines:
- Validate pod counts, data retention, and report volume before deployment.
- Increase database and cache sizes when user volume, reporting load, or retention increases.
The documentation team might draft this page using a large language model. A technical writer reviews every page and verifies it against the product before publishing it. A human makes the final decision.