Skip to content

Latest commit

 

History

History
159 lines (140 loc) · 7.42 KB

File metadata and controls

159 lines (140 loc) · 7.42 KB

../common/generic-attributes.adoc = Deploying and Installing {productname} :docinfo: :doctype: book :experimental:

WHAT?

{productname} is a multi-layered enterprise AI solution that consists of several software components and dependencies.

WHY?

Use this guide to deploy and install {productname} and its dependencies in a consistent and supported manner.

EFFORT

It takes less than one hour of reading time and requires advanced knowledge of Linux deployment.

GOAL

By the end of this guide, you will have {productname} and all required dependencies installed and ready for use.

{productname} is a versatile product consisting of multiple software layers and components. This document outlines the complete workflow for deployment and installation of all {productname} dependencies, as well as {productname} itself. You can also find references to recommended hardware and software requirements, as well as steps to take after the product installation.

💡
Hardware and software requirements

For hardware, software and application-specific requirements, refer to {productname}requirements.

../glues/ai-library-intro.adoc :override-title: Installation procedure ../tasks/AI-deployment-ailibrary-installing.adoc ../tasks/cert-manager-installing.adoc ../tasks/opensearch-installing.adoc ../tasks/milvus-installing.adoc ../tasks/ollama-installing.adoc ../references/ollama-helmchart.adoc ../tasks/owui-installing.adoc ../references/owui-helm-overrides.adoc ../references/owui-helmchart.adoc ../tasks/vllm-installing.adoc ../references/vllm-helm-overrides.adoc ../tasks/mcpo-installing.adoc ../tasks/pytorch-installing.adoc ../references/pytorch-helm-overrides.adoc ../references/pytorch-helmchart.adoc ../tasks/qdrant-installing.adoc ../references/qdrant-helm-overrides.adoc ../tasks/litellm-installing.adoc ../references/litellm-helm-overrides.adoc ../references/litellm-helmchart.adoc ../tasks/mlflow-installing.adoc ../tasks/kubeflow-installing.adoc ../tasks/kubeflow-accessing.adoc :override-title: Configuration scenarios ../references/kubeflow-configuration-scenarios.adoc :override-title: Hardening for production use ../tasks/kubeflow-hardening.adoc :override-title: Managing user profiles and namespaces ../tasks/kubeflow-user-profiles.adoc :override-title: Upgrade notes ../references/kubeflow-upgrade-notes.adoc :override-title: Known limitations ../references/kubeflow-limitations.adoc :override-title: Troubleshooting ../references/kubeflow-troubleshooting.adoc ../tasks/ai-library-apps-verifying.adoc

Alternative deployments

Selected parts of {productname} deployment can be automated or performed in a Web UI environment instead of the command line. This includes the following methods:

  • Preparing and configuring {suse}-based servers for AI workloads. This includes deployment of {rke2a}, {ranchera} or {rancherprime}, and {nvidia} GPU driver and {nvoperator} using {sainodeinstaller}.

  • Deploying {productname} workloads using {saideployer}.

  • Deploying {productname} workloads using {sailifecyclemanager}.

../concepts/ai-node-installer.adoc :override-abstract: ../tasks/ai-node-installer.adoc :override-title: {ailibrary} deployer ../tasks/ai-deployment-ailibrary-deployer.adoc :override-title: Lifecycle manager ../concepts/ai-lifecycle-manager.adoc :override-title: Installation using {ranchera} UI :override-abstract: ../tasks/ai-lifecycle-manager-installing-rancher-ui.adoc :override-title: Installation using {saioperator} :override-abstract: ../tasks/ai-lifecycle-manager-installing-ai-operator.adoc :override-title: ClusterRepo repository :override-abstract: ../tasks/ai-lifecycle-manager-clusterrepo-creating.adoc :override-title: Installing and managing applications :override-abstract: ../tasks/ai-lifecycle-manager-installing-applications.adoc