|Index|SUSE AI Factory with NVIDIA Requirements|Application-specific requirements
Applies to SUSE AI Factory with NVIDIA

3 Application-specific requirements

We recommend running each application on nodes that meet or exceed the corresponding hardware requirements.

3.1 SUSE Rancher Prime requirements

3.1.1 Supported Rancher version

SUSE AI Factory with NVIDIA requires Rancher version 2.14.x.

3.1.2 Minimum hardware requirements

Nodes for HA setup

At least 3 nodes.

RAM

A minimum of 32 GB of RAM.

CPU

At least 8 CPU cores.

Disk space

At least 200 GB of storage, preferably SSD.

3.1.3 For more information

For more detailed recommendations, refer to the following official documentation:

3.2 SUSE Security requirements

3.2.1 Minimum hardware requirements

Nodes for HA setup

The following container instances run on existing cluster nodes:

  • 1 Manager instance

  • 3 Controller instances

  • 1 Enforcer instance on each cluster node

  • 2 Scanner & Updater instances

RAM

A minimum of 2 GB of RAM.

CPU

At least 2 CPU cores.

Disk space

At least 5 GB of storage, preferably SSD.

3.2.2 For more information

For more detailed recommendations, refer to the following official documentation:

3.3 SUSE Observability requirements

3.3.1 Minimum hardware requirements

Nodes for HA setup

At least 3 nodes.

RAM

A minimum of 32 GB of RAM.

CPU

At least 16 CPU cores.

Disk space

At least 5 GB of storage, preferably SSD.

3.3.2 For more information

For more detailed recommendations, refer to the following official documentation:

3.4 NVIDIA requirements

Deploying AI applications on GPU-enabled clusters requires specific NVIDIA operators. These components manage your containerized workloads and optimize hardware performance.

3.4.1 What are the NVIDIA operator requirements?

NIM Operator

Required for deploying NIM-based applications and RAG blueprints. For more details, refer to the NVIDIA NIM operator documentation.

Network Operator

(Optional) Provides advanced networking features to optimize GPU communication and utilization across the cluster. For more details, refer to the NVIDIA network operator documentation.