> ## Documentation Index
> Fetch the complete documentation index at: https://docs.tensormesh.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Getting Started

> Pick an install method for the Tensormesh Operator and verify your cluster is ready.

The Tensormesh Operator (TMO) is a Kubernetes controller that reconciles `LMCacheEngine`
custom resources into a DaemonSet of LMCache cache servers — one per GPU node — that back
vLLM (or any inference runtime) with shared KV cache reuse.

This page covers what you need before installing and helps you pick an install method.
Each method ends up deploying the same operator and the same CRD; the difference is how
you manage the release.

## What gets installed

<CardGroup cols={3}>
  <Card title="Operator" icon="gear">
    A small controller-manager Deployment that watches `LMCacheEngine` resources cluster-wide.
  </Card>

  <Card title="CRD" icon="file-code">
    `LMCacheEngine` (`lmcache.lmcache.ai/v1alpha1`) — the spec the operator reconciles.
  </Card>

  <Card title="Engine DaemonSet" icon="layer-group">
    One LMCache cache server pod per GPU node, created by the operator from your CR.
  </Card>
</CardGroup>

## Prerequisites

<AccordionGroup>
  <Accordion title="Kubernetes 1.28 or newer" icon="dharmachakra">
    The chart's `kubeVersion` constraint is `>=1.28.0-0`. OpenShift 4.16+ and EKS / GKE / AKS
    on a supported control plane all qualify.
  </Accordion>

  <Accordion title="At least one GPU node" icon="microchip">
    The engine DaemonSet schedules onto nodes labeled `nvidia.com/gpu.present=true`. On
    clusters with the NVIDIA GPU Operator installed, this label is set automatically. On
    clusters without it, label your GPU nodes manually:

    ```bash theme={null}
    kubectl label node <node-name> nvidia.com/gpu.present=true
    ```

    On a cluster with no GPU nodes the install still succeeds — the DaemonSet stays at
    `0` desired replicas until a matching node appears.
  </Accordion>

  <Accordion title="CDI enabled (NVIDIA GPU Operator < v25.10.0)" icon="microchip">
    The NVIDIA GPU Operator exposes GPUs via the **Container Device Interface (CDI)**.
    Before GPU Operator **v25.10.0**, CDI was **off by default** (`cdi.enabled: false`) — and
    with it off, the engine/vLLM pods get **stuck in `ContainerCreating`** (no clear error).
    On those versions, enable CDI in the GPU `ClusterPolicy`:

    ```bash theme={null}
    oc patch clusterpolicy gpu-cluster-policy --type=merge -p '{
      "spec": { "cdi": { "enabled": true, "default": true } }
    }'
    ```

    GPU Operator v25.10.0+ enables CDI by default — no action needed. See
    [Troubleshooting → Pod stuck in `ContainerCreating`](/operator/installation/troubleshooting#pod-stuck-in-containercreating-cdi)
    if you hit this.
  </Accordion>

  <Accordion title="Cluster-admin permissions" icon="shield-halved">
    Installation creates cluster-scoped objects (a CRD, a `ClusterRole`, and a
    `ClusterRoleBinding`). You need permission to create those.
  </Accordion>

  <Accordion title="A CLI for your platform" icon="terminal">
    `kubectl` on Kubernetes, `oc` on OpenShift. The two are interchangeable for everything
    in these docs — examples are shown with `kubectl` unless an OpenShift-specific command
    is required.
  </Accordion>
</AccordionGroup>

## Pick an install method

<CardGroup cols={2}>
  <Card title="Helm" icon="ship-wheel" href="/operator/installation/helm">
    The primary path. Single `helm install` command, simple upgrades and rollbacks.
  </Card>

  <Card title="Modify Existing vLLM Deployment" icon="wrench" href="/operator/installation/existing-deployment">
    Patch an existing vLLM Deployment to mount the engine connection ConfigMap and use
    LMCache MP mode safely.
  </Card>

  <Card title="GitOps (Argo CD)" icon="code-branch" href="/operator/installation/gitops">
    Declarative, Git-driven installs across many clusters. Best for fleets and audited
    environments.
  </Card>

  <Card title="OpenShift" icon="circle-nodes" href="/operator/installation/openshift">
    OpenShift adds one extra flag for the privileged SCC required by `hostIPC`. Read this
    after picking Helm or GitOps if your target is OpenShift.
  </Card>

  <Card title="Kustomize" icon="puzzle-piece" href="/operator/installation/kustomize">
    Render the chart once with `helm template` and patch it with Kustomize overlays. For
    teams that standardize on Kustomize as their final renderer.
  </Card>
</CardGroup>

## Verify your cluster is ready

Run these three preflight checks before installing.

<Steps>
  <Step title="Kubernetes is 1.28 or newer">
    ```bash theme={null}
    kubectl version
    ```

    Look at the `Server Version` line — anything `v1.28.0` or newer qualifies. OpenShift
    users can also run `oc version`; the `Kubernetes Version` line is what matters.
  </Step>

  <Step title="At least one GPU node is labeled (or you accept an empty DaemonSet)">
    ```bash theme={null}
    kubectl get nodes -l nvidia.com/gpu.present=true
    ```

    A non-empty list means the engine DaemonSet has somewhere to schedule. An empty list
    is fine for an operator-only install — see
    [Install modes](/operator/installation/helm#install-modes).
  </Step>

  <Step title="You can create cluster-scoped resources">
    ```bash theme={null}
    kubectl auth can-i create customresourcedefinitions.apiextensions.k8s.io
    kubectl auth can-i create clusterroles.rbac.authorization.k8s.io
    kubectl auth can-i create clusterrolebindings.rbac.authorization.k8s.io
    ```

    All three must return `yes`. As a shortcut,
    `kubectl auth can-i '*' '*' --all-namespaces` returning `yes` means you're effectively
    cluster-admin.

    Each command also prints a `Warning: resource '...' is not namespace scoped` line —
    that's expected (these *are* cluster-scoped resources) and harmless. The `yes`/`no`
    answer is what matters.
  </Step>
</Steps>

<Note>
  If any check fails: bump the cluster version, label the GPU nodes (see the prereqs
  accordion above), or ask a cluster admin for the missing RBAC.
</Note>

## Next steps

<CardGroup cols={2}>
  <Card title="Install with Helm" icon="ship-wheel" href="/operator/installation/helm">
    Five-minute install with `helm install`.
  </Card>

  <Card title="Modify Existing vLLM Deployment" icon="wrench" href="/operator/installation/existing-deployment">
    Patch the workload you already run instead of using the demo deployment.
  </Card>

  <Card title="Troubleshooting" icon="stethoscope" href="/operator/installation/troubleshooting">
    Common install and day-2 issues.
  </Card>
</CardGroup>