docs: How to add worker pool to separate facility

docs/how-to-guides/worker-pool-separate-facility.md

@@ -0,0 +1,62 @@
+---
+title: Add worker pool in different facility on Equinix Metal
+weight: 10
+---
+
+## Introduction
+
+Equinix Metal (EM) supports inter-facility network connectivity. Building on that Lokomotive
+supports adding worker pool to a Lokomotive cluster in a different facility. The reasons to add a
+worker pool in a separate pool could be numerous viz. facility-wide HA, node type availability,
+proximity to the application users, etc.
+
+This document provides a step by step guide on adding a worker-pool to existing Lokomotive cluster
+but in a different facility than the control plane.
+
+## Prerequisites
+
+* A Lokomotive cluster accessible via `kubectl` deployed on a supported provider.
+
+* Access to Equinix Metal console and permissions to edit project-level settings.
+
+## Steps
+
+### Step 1: Enable "Backend Transfer"
+
+Go to the Equinix Metal console of your project and enable "Backend Transfer" on it, follow [this
+document](https://metal.equinix.com/developers/docs/networking/features/#backend-transfer) for
+detailed information.
+
+### Step 2: Add private CIDR of the new facility
+
+Go to **Equinix Metal console** > **IPs & Networks** > **IPs**.
+
+Now spot the _"Management"_ IP block (CIDR) for the facility of your choice and make a note of the
+`10.xx.xx.xx/25` range.
+
+Open your Lokomotive cluster's `lokocfg` file and add it to the existing `node_private_cidrs` list.
+
+```tf
+node_private_cidrs = ["10.10.10.128/25", "10.xx.xx.xx/25"]
+```
+### Step 3: Add worker pool with a different facility
+
+Add the following snippet to your existing `lokocfg` file, under the `cluster "packet"` section:
+
+```tf
+  worker_pool "worker-new-facility" {
+    count    = 1
+    facility = "<new facility>"
+  }
+```
+
+### Step 4: Apply changes
+
+Execute the following command to apply the above changes:
+
+```
+lokoctl cluster apply -v --skip-components
+```
+
+Once the above command is successfully executed, you will have a worker pool in a separate facility
+connected to your existing cluster.