Add baseline configuration for performance comparison of llm-d

skills/compare-llm-d-configurations/SKILL.md

@@ -93,6 +93,51 @@ Only needed if at least one run is being deployed fresh. Both runs share the sam
 2. Check for an active `oc project`: `oc project -q 2>/dev/null`
 3. Otherwise ask the user.
 
+### 0.5 Check for Baseline Configuration
+
+If either run is labeled as "baseline" (doesnt use llm-d scheduling) or the user indicates one configuration is a baseline run, perform the following checks:
+
+1. **Verify baseline service exists**:
+   ```bash
+   kubectl get svc llm-d-baseline-model-server -n $NAMESPACE
+   ```
+
+2. **If the service does not exist**, inform the user that you will create it, then apply the baseline service YAML:
+   ```bash
+   kubectl apply -f skills/compare-llm-d-configurations/llm-d-baseline-model-server-svc.yaml -n $NAMESPACE
+   ```
+
+   Or if that file is not available locally, create it inline:
+   ```bash
+   cat <<EOF | kubectl apply -n $NAMESPACE -f -
+   apiVersion: v1
+   kind: Service
+   metadata:
+     name: llm-d-baseline-model-server
+   spec:
+     selector:
+       llm-d.ai/inference-serving: "true"
+     ports:
+     - name: http
+       protocol: TCP
+       port: 8000
+       targetPort: 8000
+     type: ClusterIP
+   EOF
+   ```
+
+3. **Verify the service has endpoints**:
+   ```bash
+   kubectl get endpoints llm-d-baseline-model-server -n $NAMESPACE
+   ```
+
+   If no endpoints exist, inform the user that pods with the label `llm-d.ai/inference-serving=true` must be running for the baseline service to work. Check the running pod labels, list them to the user and suggest which label to use for the sellector (update the baseline service accordingly).
+
+4. **Set baseline base_url**: When running the benchmark for the baseline configuration (as Run A or as Run B), ensure the Run's config.yaml uses:
+   ```yaml
+   base_url: http://llm-d-baseline-model-server.<namespace>.svc.cluster.local:8000
+   ```
+
 ---
 
 ## Phase 1: Run A

skills/compare-llm-d-configurations/llm-d-baseline-model-server-svc.yaml

@@ -0,0 +1,13 @@
+apiVersion: v1
+kind: Service
+metadata:
+  name: llm-d-baseline-model-server
+spec:
+  selector:
+    llm-d.ai/inference-serving: "true"
+  ports:
+  - name: http
+    protocol: TCP
+    port: 8000
+    targetPort: 8000
+  type: ClusterIP