Skip to content

infrastructure_components_guide.md

Latest commit

 

History

History
78 lines (56 loc) · 5.18 KB

infrastructure_components_guide.md

File metadata and controls

78 lines (56 loc) · 5.18 KB
Raw

Infrastructure Components Guide

This document describes how to install infrastructure components on a K8s cluster.

All instructions need to be executed on your local machine from the deployment folder.

If you are using your own custom Kubernetes cluster, you may need to install additional infrastructure components before deploying Intel® AI for Enterprise RAG application. These include the NFS server for shared storage, the Gaudi operator (for Habana Gaudi AI accelerator support), Intel GPU device plugin (for Intel® Arc™ B-Series experimental support), Velero, local registry, or other supported services.

To install additional infrastructure components to your cluster:

  1. Edit the configuration file:
    • Open inventory/test-cluster/config.yaml.
    • Fill in the required values:
      • deploy_k8s: false
      • gaudi_operator: true set to true only if you are working with Gaudi nodes and want to install the Gaudi software stack via operator.
    • is_bmg_platform_enable: true for Intel® Arc™ B-Series deployments (experimental, testing purposes only).
    • intel_gpu_plugin: true when Intel GPU plugin should be installed/updated on the cluster.
    • intel_gpu_plugin_version: set required Intel GPU plugin version for your environment.
      • install_csi - set one of the following options:
        • local-path-provisioner for single-node deployment
        • nfs for multi-node deployment; when choosing this option, fill in the nfs section in config.yaml
        • netapp-trident: Use for enterprise deployments with NetApp ONTAP storage backend; when choosing this option, fill in the netapp-trident section in config.yaml
      • local_registry: Default is false. Set to true only if you have a multi-node setup and you don't want to use images from public registry. When enabled, also configure:
        • insecure_registry: "<node-name>:32000" where <node-name> is the Kubernetes node name where the registry pod will be deployed
      • httpProxy and httpsProxy values if you are using a proxy.
      • kubeconfig: path to your kubeconfig
      • velero: true if you want to install the Velero backup tool.

Note

The pre-install tag automatically configures system limits (file descriptors, process limits, inotify watches, and kernel parameters) on all cluster nodes. These configurations are essential for optimal Enterprise RAG performance and are applied before the main installation process.

Note

The local registry option creates a Kubernetes pod with registry functionality and configures Docker and containerd settings to enable pushing and pulling images to the registry pod. This is particularly useful for multi-node clusters where a standard Docker registry would only be accessible from a single node.

  1. Validate hardware resources and config.yaml:

    ansible-playbook playbooks/validate.yaml --tags hardware,config -i inventory/test-cluster/inventory.ini -e @inventory/test-cluster/config.yaml

Note

If this is a Gaudi deployment, add the flag -e is_gaudi_platform=true.

Note

If this is an Intel® Arc™ B-Series deployment, add the flag -e is_bmg_platform=true (experimental, for testing purposes only).

  1. Configure system limits (recommended before installation):

    ansible-playbook -K playbooks/infrastructure.yaml --tags pre-install -i inventory/test-cluster/inventory.ini -e @inventory/test-cluster/config.yaml

    This will configure system limits on all cluster nodes including file descriptors, process limits, inotify watches, and kernel parameters required for Enterprise RAG.

  2. Install infrastructure components (NFS, Gaudi operator, local registry, or others):

    ansible-playbook -K playbooks/infrastructure.yaml --tags post-install -i inventory/test-cluster/inventory.ini -e @inventory/test-cluster/config.yaml

    This will install and configure the NFS server, Gaudi operator, local registry, or Velero as specified in your configuration.

For Intel® Arc™ B-Series deployments (experimental, testing purposes only), append -e is_bmg_platform=true.

If you also need to provision Kubernetes (optional), run infrastructure with full install tags instead of component-only mode:

ansible-playbook -K playbooks/infrastructure.yaml --tags configure,install -i inventory/test-cluster/inventory.ini -e @inventory/test-cluster/config.yaml -e is_bmg_platform=true

Note

You can enable several components in the same run if multiple components are needed. Additional components may be supported via post-install in the future.

Velero Backup Storage

Velero requires a Backup Storage Location (BSL) — an S3-compatible object store — to hold backup metadata and volume-snapshot data. Two modes are supported:

  • Local (default): An in-cluster SeaweedFS instance is deployed alongside Velero automatically. Suitable for evaluation; backup data does not survive cluster loss.
  • External: Velero connects to any S3-compatible service outside the cluster (AWS S3, MinIO, Ceph, etc.). Required for real disaster-recovery scenarios.

For a full comparison and configuration reference, see docs/backup_storage_configuration.md.