Introduction to SUSE AI deployment
- WHAT?
Basic information about SUSE AI deployment workflow.
- WHY?
To better understand the SUSE AI deployment process.
- EFFORT
Less than 15 minutes of reading and a basic knowledge of Linux deployment.
1 Deployment overview #
This topic outlines the complete workflow to deploy the SUSE AI product. It aims to clarify the deployment process to get good understanding of its complexity, find minimum hardware requirements, and steps to take after the deployment.
1.1 Requirements #
Before deploying SUSE AI, the following must be fulfilled:
You have purchased Rancher Prime entitlement.
You have installed Rancher Manager as documented in https://ranchermanager.docs.rancher.com/getting-started/installation-and-upgrade/install-upgrade-on-a-kubernetes-cluster.
You have deployed and configured SUSE Security as documented in https://open-docs.neuvector.com/.
You have deployed and configured SUSE Observability as documented in https://docs.stackstate.com/.
For basic functionality, SUSE AI has the same hardware requirements as the RKE2 distribution with at least 32 GB of RAM. Refer to https://docs.rke2.io/install/requirements for details.
To use the NVIDIA GPU Operator, you need a compatible GPU hardware. Refer to https://docs.nvidia.com/datacenter/cloud-native/gpu-operator/latest/platform-support.html for details.
1.2 Cluster preparation #
Install and register SUSE Linux Micro 6.0 or later on each RKE2 cluster node. Refer to https://documentation.suse.com/sle-micro/6.0/ for details.
(Optional) If you plan to utilize the computing power of NVIDIA GPUs, install the NVIDIA GPU driver on cluster nodes with GPUs. Refer to https://documentation.suse.com/suse-ai/1.0/html/NVIDIA-GPU-driver-on-SL-Micro/index.html for details.
(Optional) Install the NVIDIAcontainer toolkit on GPU enabled systems. Refer to https://documentation.suse.com/suse-ai/1.0/html/NVIDIA-GPU-driver-on-SL-Micro/index.html#nvidia-gpu-validation for details.
Install RKE2 Kubernetes distribution on the cluster nodes. Refer to https://docs.rke2.io/ for details.
Connect the RKE2 cluster to Rancher Manager.
Configure the GPU enabled nodes so that the SUSE AI containers are assigned to Pods that run on nodes equipped with NVIDIA GPU hardware. Find more details assigning Pods to nodes in https://kubernetes.io/docs/concepts/scheduling-eviction/assign-pod-node/.
(Optional) If you previously installed the NVIDIA GPU driver on hosts with GPU hardware, install NVIDIA GPU Operator with the additional option
--set driver.enabled=false. Refer to https://documentation.suse.com/suse-ai/1.0/html/NVIDIA-Operator-installation/index.html.(Optional) Configure SUSE Security to scan the nodes that will be used for SUSE AI. Although this step is not required, we strongly encourage it to ensure the security in production environment.
Configure SUSE Observability to observe the nodes with the SUSE AI application.
2 Legal Notice #
Copyright© 2006–2024 SUSE LLC and contributors. All rights reserved.
Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or (at your option) version 1.3; with the Invariant Section being this copyright notice and license. A copy of the license version 1.2 is included in the section entitled “GNU Free Documentation License”.
For SUSE trademarks, see https://www.suse.com/company/legal/. All other third-party trademarks are the property of their respective owners. Trademark symbols (®, ™ etc.) denote trademarks of SUSE and its affiliates. Asterisks (*) denote third-party trademarks.
All information found in this book has been compiled with utmost attention to detail. However, this does not guarantee complete accuracy. Neither SUSE LLC, its affiliates, the authors, nor the translators shall be held liable for possible errors or the consequences thereof.