Additional edits and updates

Author: AMD-melliott

docs/advanced/health-checks.md

@@ -6,8 +6,7 @@ The AMD GPU Device Plugin for Kubernetes enables the use of AMD GPUs as schedula
 
 - Implements the Kubernetes Device Plugin API for AMD GPUs
 - Exposes AMD GPUs as `amd.com/gpu` resources in Kubernetes
-- Supports automated node labeling with GPU-specific information
-- Provides optional health monitoring for GPUs
+- Provides automated node labeling with detailed GPU properties (device ID, VRAM, compute units, etc.)
 - Enables fine-grained GPU allocation for containers
 
 ## System Requirements
@@ -20,8 +19,6 @@ See the [ROCm System Requirements](https://rocm.docs.amd.com/projects/install-on
 
 ## Quick Start
 
-Deploy the device plugin on your Kubernetes cluster:
-
 To deploy the device plugin, run it on all nodes equipped with AMD GPUs. The simplest way to do this is by creating a Kubernetes [DaemonSet](https://kubernetes.io/docs/concepts/workloads/controllers/daemonset/). A pre-built Docker image is available on [DockerHub](https://hub.docker.com/r/rocm/k8s-device-plugin), and a predefined YAML file named [k8s-ds-amdgpu-dp.yaml](https://raw.githubusercontent.com/ROCm/k8s-device-plugin/master/k8s-ds-amdgpu-dp.yaml) is included in this repository.
 
 Create a DaemonSet in your Kubernetes cluster with the following command:
@@ -36,20 +33,34 @@ Alternatively, you can pull directly from the web:
 kubectl create -f https://raw.githubusercontent.com/ROCm/k8s-device-plugin/master/k8s-ds-amdgpu-dp.yaml
 ```
 
-To enable the experimental device health check, use `k8s-ds-amdgpu-dp-health.yaml` **after** setting `--allow-privileged=true` for kube-apiserver.
+### Deploy the Node Labeler (Optional)
 
-## Example Workload
+For enhanced GPU discovery and scheduling, deploy the AMD GPU Node Labeler:
+
+```bash
+kubectl create -f k8s-ds-amdgpu-labeller.yaml
+```
+
+This will automatically label nodes with GPU-specific information such as VRAM size, compute units, and device IDs.
 
-You can restrict workloads to a node with a GPU by adding `resources.limits` to the pod definition. An example pod definition is provided in `example/pod/alexnet-gpu.yaml`. Create the pod by running:
+### Verify Installation
+
+After deploying the device plugin, verify that your AMD GPUs are properly recognized as schedulable resources:
 
 ```bash
-kubectl create -f alexnet-gpu.yaml
+# List all nodes with their AMD GPU capacity
+kubectl get nodes -o custom-columns=NAME:.metadata.name,GPU:"status.capacity.amd\.com/gpu"
+
+NAME             GPU
+k8s-node-01      8
 ```
 
-or
+## Example Workload
+
+You can restrict workloads to a node with a GPU by adding `resources.limits` to the pod definition. An example pod definition is provided in [example/pod/pytorch.yam](https://raw.githubusercontent.com/ROCm/k8s-device-plugin/master/example/pod/pytorch.yaml)l. Create the pod by running:
 
 ```bash
-kubectl create -f https://raw.githubusercontent.com/ROCm/k8s-device-plugin/master/example/pod/tensorflow-gpu.yaml
+kubectl create -f https://raw.githubusercontent.com/ROCm/k8s-device-plugin/master/example/pod/pytorch.yaml
 ```
 
 Check the pod status with:
@@ -61,13 +72,9 @@ kubectl describe pods
 After the pod is running, view the benchmark results with:
 
 ```bash
-kubectl logs rocm-test-pod rocm-test-container
+kubectl pytorch-gpu-pod-example
 ```
 
-## Health Checks
-
-This plugin extends more granular health detection per GPU using the exporter health service over a gRPC socket service mounted on `/var/lib/amd-metrics-exporter/`.
-
 ## Contributing
 
-We welcome contributions to this project! Please refer to the [Contributing Guide](contributing/development.md) for details on how to get involved.
+We welcome contributions to this project! Please refer to the [Development Guidelines](contributing/development.md) for details on how to get involved.

docs/advanced/monitoring.md

@@ -8,11 +8,6 @@ subtrees:
   - file: user-guide/installation
   - file: user-guide/configuration
   - file: user-guide/examples
-- caption: Advanced
-  entries:
-  - file: advanced/health-checks
-  - file: advanced/monitoring
-  - file: advanced/node-labelling
 - caption: Contributing
   entries:
   - file: contributing/development