update inferno metrics naming to wva

Signed-off-by: Mohammed Abdi <mohammed.munir.abdi@ibm.com>

Author: mamy-CS

CONTRIBUTING.md

@@ -256,7 +256,7 @@ kubectl logs -n workload-variant-autoscaler-system \
 kubectl describe variantautoscaling <name> -n <namespace>
 
 # Check emitted metrics
-kubectl get --raw "/apis/external.metrics.k8s.io/v1beta1/namespaces/<namespace>/inferno_desired_replicas" | jq
+kubectl get --raw "/apis/external.metrics.k8s.io/v1beta1/namespaces/<namespace>/wva_desired_replicas" | jq
 ```
 
 ### Cleaning Up

charts/workload-variant-autoscaler/README.md

@@ -300,5 +300,5 @@ kubectl logs pod prometheus-adapter-xxxxx -n openshift-user-workload-monitoring
 ```
 3. Check, after a few minutes following installation, for metric collection
 ```
-kubectl get --raw "/apis/external.metrics.k8s.io/v1beta1/namespaces/$NAMESPACE/inferno_desired_replicas" | jq
+kubectl get --raw "/apis/external.metrics.k8s.io/v1beta1/namespaces/$NAMESPACE/wva_desired_replicas" | jq
 ```

charts/workload-variant-autoscaler/templates/hpa.yaml

@@ -36,7 +36,7 @@ spec:
   - type: External
     external:
       metric:
-        name: inferno_desired_replicas
+        name: wva_desired_replicas
         selector:
           matchLabels:
             variant_name: {{ printf "%s-decode" .Values.llmd.modelName }}

config/samples/hpa-integration.yaml

@@ -27,7 +27,7 @@ spec:
   - type: External
     external:
       metric:
-        name: inferno_desired_replicas
+        name: wva_desired_replicas
         selector:
           matchLabels:
             variant_name: vllme-deployment

config/samples/prometheus-adapter-values-ocp.yaml

@@ -4,15 +4,15 @@ prometheus:
 
 rules:
   external:
-  - seriesQuery: 'inferno_desired_replicas{variant_name!="",exported_namespace!=""}'
+  - seriesQuery: 'wva_desired_replicas{variant_name!="",exported_namespace!=""}'
     resources:
       overrides:
         exported_namespace: {resource: "namespace"}
         variant_name: {resource: "deployment"}
     name:
-      matches: "^inferno_desired_replicas"
-      as: "inferno_desired_replicas"
-    metricsQuery: 'inferno_desired_replicas{<<.LabelMatchers>>}'
+      matches: "^wva_desired_replicas"
+      as: "wva_desired_replicas"
+    metricsQuery: 'wva_desired_replicas{<<.LabelMatchers>>}'
 
 replicas: 2
 logLevel: 4

config/samples/prometheus-adapter-values.yaml

@@ -4,15 +4,15 @@ prometheus:
 
 rules:
   external:
-  - seriesQuery: 'inferno_desired_replicas{variant_name!="",exported_namespace!=""}'
+  - seriesQuery: 'wva_desired_replicas{variant_name!="",exported_namespace!=""}'
     resources:
       overrides:
         exported_namespace: {resource: "namespace"}
         variant_name: {resource: "deployment"}  
     name:
-      matches: "^inferno_desired_replicas"
-      as: "inferno_desired_replicas"
-    metricsQuery: 'inferno_desired_replicas{<<.LabelMatchers>>}'
+      matches: "^wva_desired_replicas"
+      as: "wva_desired_replicas"
+    metricsQuery: 'wva_desired_replicas{<<.LabelMatchers>>}'
 
 replicas: 2
 logLevel: 4

deploy/README.md

@@ -538,7 +538,7 @@ spec:
   - type: External
     external:
       metric:
-        name: inferno_desired_replicas
+        name: wva_desired_replicas
         selector:
           matchLabels:
             variant_name: my-vllm-deployment-decode
@@ -717,7 +717,7 @@ kubectl port-forward -n <monitoring-namespace> svc/prometheus-k8s 9090:9090
 kubectl logs -n workload-variant-autoscaler-system -l app.kubernetes.io/name=workload-variant-autoscaler | grep "Collected metrics"
 
 # 4. Verify external metrics API (if using HPA)
-kubectl get --raw "/apis/external.metrics.k8s.io/v1beta1/namespaces/<namespace>/inferno_desired_replicas" | jq
+kubectl get --raw "/apis/external.metrics.k8s.io/v1beta1/namespaces/<namespace>/wva_desired_replicas" | jq
 ```
 
 ### Monitoring WVA
@@ -807,7 +807,7 @@ watch kubectl get hpa -n <namespace>
 watch kubectl get pods -n <namespace>
 
 # Watch external metrics
-watch 'kubectl get --raw "/apis/external.metrics.k8s.io/v1beta1/namespaces/<namespace>/inferno_desired_replicas" | jq'
+watch 'kubectl get --raw "/apis/external.metrics.k8s.io/v1beta1/namespaces/<namespace>/wva_desired_replicas" | jq'
 ```
 
 ## Troubleshooting
@@ -913,14 +913,14 @@ my-hpa     Deployment/vllm    <unknown>/1(avg)   1         10        1
 kubectl describe hpa <name> -n <namespace>
 
 # Check external metrics API on the specified namespace
-kubectl get --raw "/apis/external.metrics.k8s.io/v1beta1/namespaces/<your-namespace>/inferno_desired_replicas" | jq
+kubectl get --raw "/apis/external.metrics.k8s.io/v1beta1/namespaces/<your-namespace>/wva_desired_replicas" | jq
 
 # Check Prometheus Adapter logs
 kubectl logs -n <monitoring-namespace> deployment/prometheus-adapter
 
 # Check if WVA is emitting the metric
 kubectl logs -n workload-variant-autoscaler-system -l app.kubernetes.io/name=workload-variant-autoscaler | \
-  grep "inferno_desired_replicas"
+  grep "wva_desired_replicas"
 ```
 
 **Common causes**:
@@ -941,7 +941,7 @@ kubectl get configmap prometheus-adapter -n <monitoring-namespace> -o yaml
 
 # Verify metric exists in Prometheus
 kubectl port-forward -n <monitoring-namespace> svc/prometheus-k8s 9090:9090
-# Query: inferno_desired_replicas{variant_name="<name>"}
+# Query: wva_desired_replicas{variant_name="<name>"}
 ```
 
 ### Getting Help

deploy/install.sh

@@ -732,7 +732,7 @@ print_summary() {
     echo "   kubectl logs -n $WVA_NS -l app.kubernetes.io/name=workload-variant-autoscaler -f"
     echo ""
     echo "4. Check external metrics API:"
-    echo "   kubectl get --raw \"/apis/external.metrics.k8s.io/v1beta1/namespaces/$LLMD_NS/inferno_desired_replicas\" | jq"
+    echo "   kubectl get --raw \"/apis/external.metrics.k8s.io/v1beta1/namespaces/$LLMD_NS/wva_desired_replicas\" | jq"
     echo ""
     echo "5. Port-forward Prometheus to view metrics:"
     echo "   kubectl port-forward -n $MONITORING_NAMESPACE svc/${PROMETHEUS_SVC_NAME} ${PROMETHEUS_PORT}:${PROMETHEUS_PORT}"

deploy/kubernetes/README.md

@@ -445,7 +445,7 @@ kubectl describe variantautoscaling ms-inference-scheduling-llm-d-modelservice-d
 kubectl get hpa -n llm-d-inference-scheduling
 
 # Check external metrics
-kubectl get --raw "/apis/external.metrics.k8s.io/v1beta1/namespaces/llm-d-inference-scheduling/inferno_desired_replicas" | jq
+kubectl get --raw "/apis/external.metrics.k8s.io/v1beta1/namespaces/llm-d-inference-scheduling/wva_desired_replicas" | jq
 ```
 
 ### Monitor WVA Logs (See Metrics Validation!)
@@ -470,7 +470,7 @@ kubectl port-forward -n workload-variant-autoscaler-monitoring \
 
 # Visit http://localhost:9090
 # Query: vllm:request_success_total
-# Query: inferno_desired_replicas
+# Query: wva_desired_replicas
 ```
 
 ### Access Grafana Dashboards

deploy/openshift/README.md

@@ -311,7 +311,7 @@ export HF_TOKEN="hf_xxxxxxxxxxxxxxxxxxxxx"
 
 ```bash
 kubectl get pods -n openshift-user-workload-monitoring | grep prometheus-adapter
-kubectl get --raw "/apis/external.metrics.k8s.io/v1beta1/namespaces/llm-d-inference-scheduler/inferno_desired_replicas" | jq
+kubectl get --raw "/apis/external.metrics.k8s.io/v1beta1/namespaces/llm-d-inference-scheduler/wva_desired_replicas" | jq
 ```
 
 ### vLLM Pods Not Starting
@@ -344,7 +344,7 @@ kubectl get variantautoscaling -n llm-d-inference-scheduler
 kubectl get hpa -n llm-d-inference-scheduler
 
 # Check external metrics
-kubectl get --raw "/apis/external.metrics.k8s.io/v1beta1/namespaces/llm-d-inference-scheduler/inferno_desired_replicas" | jq
+kubectl get --raw "/apis/external.metrics.k8s.io/v1beta1/namespaces/llm-d-inference-scheduler/wva_desired_replicas" | jq
 ```
 
 ### Monitor WVA Logs