Merge branch 'llm-d-incubation:main' into statefulset

Author: meffmadd

charts/llm-d-modelservice/Chart.yaml

@@ -13,7 +13,7 @@ type: application
 # This is the chart version. This version number should be incremented each time you make changes
 # to the chart and its templates, including the app version.
 # Versions are expected to follow Semantic Versioning (https://semver.org/)
-version: "v0.4.12"
+version: "v0.4.15"
 # This is the version number of the application being deployed. This version number should be
 # incremented each time you make changes to the application. Versions are not expected to
 # follow Semantic Versioning. They should reflect the version the application is using.

charts/llm-d-modelservice/templates/_helpers.tpl

@@ -159,8 +159,8 @@ affinity:
     {{- end }}
   image: {{ required "routing.proxy.image must be specified" .proxy.image }}
   imagePullPolicy: {{ default "Always" .proxy.imagePullPolicy }}
-  env:
 {{- if and .Values.tracing .Values.tracing.enabled }}
+  env:
     - name: OTEL_SERVICE_NAME
       value: {{ .Values.tracing.serviceNames.routingProxy | quote }}
     - name: OTEL_EXPORTER_OTLP_ENDPOINT
@@ -174,7 +174,11 @@ affinity:
 {{- end }}
   ports:
     - containerPort: {{ default 8000 .servicePort }}
+  {{- if .proxy.resources }}
+  resources: {{- toYaml .proxy.resources | nindent 4 }}
+  {{- else }}
   resources: {}
+  {{- end }}
   restartPolicy: Always
   securityContext:
     allowPrivilegeEscalation: false
@@ -291,14 +295,23 @@ nvidia.com/gpu
 {{- end -}}
 {{- end }}
 
-{{/* Get accelerator environment variables based on type */}}
+{{/*
+Get accelerator environment variables based on type.
+Filter out any that are
+*/}}
 {{- define "llm-d-modelservice.acceleratorEnv" -}}
 {{- $acceleratorType := include "llm-d-modelservice.acceleratorType" . -}}
 {{- if and (ne $acceleratorType "cpu") (hasKey .Values.accelerator.env $acceleratorType) -}}
 {{- $envVars := index .Values.accelerator.env $acceleratorType -}}
+{{- $userNames := list -}}
+{{- range (default (list) .container.env) -}}
+  {{- $userNames = append $userNames .name -}}
+{{- end -}}
 {{- range $envVars }}
+{{- if not (has .name $userNames) }}
 - name: {{ .name }}
   value: {{ .value | quote }}
+{{- end }}
 {{- end -}}
 {{- end -}}
 {{- end }}

charts/llm-d-modelservice/values.schema.json

@@ -480,7 +480,7 @@
                                                     "description": "Selects a key from a ConfigMap.",
                                                     "properties": {
                                                         "key": {
-                                                            "description": "The key to select.",
+                                                            "description": "The key to select from the ConfigMap's Data field. Keys in the BinaryData field are not currently propagated to container env vars.",
                                                             "type": "string"
                                                         },
                                                         "name": {
@@ -617,7 +617,7 @@
                                     "description": "EnvFromSource represents the source of a set of ConfigMaps or Secrets",
                                     "properties": {
                                         "configMapRef": {
-                                            "description": "ConfigMapEnvSource selects a ConfigMap to populate the environment variables with.\n\nThe contents of the target ConfigMap's Data field will represent the key-value pairs as environment variables.",
+                                            "description": "ConfigMapEnvSource selects a ConfigMap to populate the environment variables with.\n\nThe contents of the target ConfigMap's Data field will represent the key-value pairs as environment variables. Keys in the BinaryData field are not currently propagated to container env vars.",
                                             "properties": {
                                                 "name": {
                                                     "description": "Name of the referent. This field is effectively required, but due to backwards compatibility is allowed to be empty. Instances of this type with an empty value here are almost certainly wrong. More info: https://kubernetes.io/docs/concepts/overview/working-with-objects/names/#names",
@@ -1707,7 +1707,7 @@
                                     "description": "VolumeMount describes a mounting of a Volume within a container.",
                                     "properties": {
                                         "mountPath": {
-                                            "description": "Path within the container at which the volume should be mounted.  Must not contain ':'.",
+                                            "description": "Path within the container at which the volume should be mounted.",
                                             "type": "string"
                                         },
                                         "mountPropagation": {
@@ -2089,7 +2089,7 @@
                                                     "description": "Selects a key from a ConfigMap.",
                                                     "properties": {
                                                         "key": {
-                                                            "description": "The key to select.",
+                                                            "description": "The key to select from the ConfigMap's Data field. Keys in the BinaryData field are not currently propagated to container env vars.",
                                                             "type": "string"
                                                         },
                                                         "name": {
@@ -2226,7 +2226,7 @@
                                     "description": "EnvFromSource represents the source of a set of ConfigMaps or Secrets",
                                     "properties": {
                                         "configMapRef": {
-                                            "description": "ConfigMapEnvSource selects a ConfigMap to populate the environment variables with.\n\nThe contents of the target ConfigMap's Data field will represent the key-value pairs as environment variables.",
+                                            "description": "ConfigMapEnvSource selects a ConfigMap to populate the environment variables with.\n\nThe contents of the target ConfigMap's Data field will represent the key-value pairs as environment variables. Keys in the BinaryData field are not currently propagated to container env vars.",
                                             "properties": {
                                                 "name": {
                                                     "description": "Name of the referent. This field is effectively required, but due to backwards compatibility is allowed to be empty. Instances of this type with an empty value here are almost certainly wrong. More info: https://kubernetes.io/docs/concepts/overview/working-with-objects/names/#names",
@@ -3316,7 +3316,7 @@
                                     "description": "VolumeMount describes a mounting of a Volume within a container.",
                                     "properties": {
                                         "mountPath": {
-                                            "description": "Path within the container at which the volume should be mounted.  Must not contain ':'.",
+                                            "description": "Path within the container at which the volume should be mounted.",
                                             "type": "string"
                                         },
                                         "mountPropagation": {
@@ -3632,6 +3632,13 @@
                             "title": "imagePullPolicy",
                             "type": "string"
                         },
+                        "resources": {
+                            "additionalProperties": true,
+                            "description": "Resource requests and limits for the routing proxy sidecar container. When not set, no resource constraints are applied (resources: {}). @schema type: object additionalProperties: true @schema",
+                            "required": [],
+                            "title": "resources",
+                            "type": "object"
+                        },
                         "secure": {
                             "default": false,
                             "description": "Boolean: adds the `--secure-proxy` flag to the routingSidecar with your chosen value.  Arg is ommitted by default for compatability with legacy sidecar images.",

charts/llm-d-modelservice/values.schema.tmpl.json

@@ -1004,6 +1004,13 @@
               "title": "imagePullPolicy",
               "type": "string"
             },
+            "resources": {
+              "additionalProperties": true,
+              "description": "Resource requests and limits for the routing proxy sidecar container. When not set, no resource constraints are applied (resources: {}). @schema type: object additionalProperties: true @schema",
+              "required": [],
+              "title": "resources",
+              "type": "object"
+            },
             "secure": {
               "default": false,
               "description": "Boolean: adds the `--secure-proxy` flag to the routingSidecar with your chosen value.  Arg is ommitted by default for compatability with legacy sidecar images.",

charts/llm-d-modelservice/values.yaml

@@ -251,6 +251,14 @@ routing:
     # @schema
     # zapTimeEncoding: epoch
 
+    # -- Resource requests and limits for the routing proxy sidecar container.
+    # When not set, no resource constraints are applied (resources: {}).
+    # @schema
+    # type: object
+    # additionalProperties: true
+    # @schema
+    resources: {}
+
 # @schema
 # additionalProperties: true
 # @schema

examples/output-cpu.yaml

@@ -6,7 +6,7 @@ kind: ServiceAccount
 metadata:
   name: cpu-sim-llm-d-modelservice
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.12
+    helm.sh/chart: llm-d-modelservice-v0.4.15
     app.kubernetes.io/version: "v0.4.0"
     app.kubernetes.io/managed-by: Helm
 ---
@@ -16,7 +16,7 @@ kind: Deployment
 metadata:
   name: cpu-sim-llm-d-modelservice-decode
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.12
+    helm.sh/chart: llm-d-modelservice-v0.4.15
     app.kubernetes.io/version: "v0.4.0"
     app.kubernetes.io/managed-by: Helm
 spec:
@@ -44,7 +44,6 @@ spec:
             - --secure-proxy=false
           image: ghcr.io/llm-d/llm-d-routing-sidecar:latest
           imagePullPolicy: Always
-          env:
           ports:
             - containerPort: 8000
           resources: {}
@@ -106,7 +105,7 @@ kind: Deployment
 metadata:
   name: cpu-sim-llm-d-modelservice-prefill
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.12
+    helm.sh/chart: llm-d-modelservice-v0.4.15
     app.kubernetes.io/version: "v0.4.0"
     app.kubernetes.io/managed-by: Helm
 spec:

examples/output-dra.yaml

@@ -6,7 +6,7 @@ kind: ServiceAccount
 metadata:
   name: dra-llm-d-modelservice
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.12
+    helm.sh/chart: llm-d-modelservice-v0.4.15
     app.kubernetes.io/version: "v0.4.0"
     app.kubernetes.io/managed-by: Helm
 ---
@@ -16,7 +16,7 @@ kind: Deployment
 metadata:
   name: dra-llm-d-modelservice-decode
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.12
+    helm.sh/chart: llm-d-modelservice-v0.4.15
     app.kubernetes.io/version: "v0.4.0"
     app.kubernetes.io/managed-by: Helm
 spec:
@@ -116,7 +116,7 @@ kind: ResourceClaimTemplate
 metadata:
   name: intel-gaudi-claim-template-decode
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.12
+    helm.sh/chart: llm-d-modelservice-v0.4.15
     app.kubernetes.io/version: "v0.4.0"
     app.kubernetes.io/managed-by: Helm
     llm-d.ai/role: decode

examples/output-gaudi.yaml

@@ -6,7 +6,7 @@ kind: ServiceAccount
 metadata:
   name: gaudi-llm-d-modelservice
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.12
+    helm.sh/chart: llm-d-modelservice-v0.4.15
     app.kubernetes.io/version: "v0.4.0"
     app.kubernetes.io/managed-by: Helm
 ---
@@ -16,7 +16,7 @@ kind: Deployment
 metadata:
   name: gaudi-llm-d-modelservice-decode
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.12
+    helm.sh/chart: llm-d-modelservice-v0.4.15
     app.kubernetes.io/version: "v0.4.0"
     app.kubernetes.io/managed-by: Helm
 spec:

examples/output-heterogeneous-pd.yaml

@@ -6,7 +6,7 @@ kind: ServiceAccount
 metadata:
   name: heterogeneous-pd-llm-d-modelservice
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.12
+    helm.sh/chart: llm-d-modelservice-v0.4.15
     app.kubernetes.io/version: "v0.4.0"
     app.kubernetes.io/managed-by: Helm
 ---
@@ -16,7 +16,7 @@ kind: Deployment
 metadata:
   name: heterogeneous-pd-llm-d-modelservice-decode
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.12
+    helm.sh/chart: llm-d-modelservice-v0.4.15
     app.kubernetes.io/version: "v0.4.0"
     app.kubernetes.io/managed-by: Helm
 spec:
@@ -44,7 +44,6 @@ spec:
             - --secure-proxy=false
           image: ghcr.io/llm-d/llm-d-routing-sidecar:latest
           imagePullPolicy: Always
-          env:
           ports:
             - containerPort: 8000
           resources: {}
@@ -132,7 +131,7 @@ kind: Deployment
 metadata:
   name: heterogeneous-pd-llm-d-modelservice-prefill
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.12
+    helm.sh/chart: llm-d-modelservice-v0.4.15
     app.kubernetes.io/version: "v0.4.0"
     app.kubernetes.io/managed-by: Helm
 spec:
@@ -225,7 +224,7 @@ kind: ResourceClaimTemplate
 metadata:
   name: nvidia-claim-template-decode
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.12
+    helm.sh/chart: llm-d-modelservice-v0.4.15
     app.kubernetes.io/version: "v0.4.0"
     app.kubernetes.io/managed-by: Helm
     llm-d.ai/role: decode

examples/output-pd-mnnvl.yaml

@@ -6,7 +6,7 @@ kind: ServiceAccount
 metadata:
   name: pd-mnnvl-llm-d-modelservice
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.12
+    helm.sh/chart: llm-d-modelservice-v0.4.15
     app.kubernetes.io/version: "v0.4.0"
     app.kubernetes.io/managed-by: Helm
 ---
@@ -16,7 +16,7 @@ kind: Deployment
 metadata:
   name: pd-mnnvl-llm-d-modelservice-decode
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.12
+    helm.sh/chart: llm-d-modelservice-v0.4.15
     app.kubernetes.io/version: "v0.4.0"
     app.kubernetes.io/managed-by: Helm
 spec:
@@ -44,7 +44,6 @@ spec:
             - --secure-proxy=false
           image: ghcr.io/llm-d/llm-d-routing-sidecar:latest
           imagePullPolicy: Always
-          env:
           ports:
             - containerPort: 8000
           resources: {}
@@ -132,7 +131,7 @@ kind: Deployment
 metadata:
   name: pd-mnnvl-llm-d-modelservice-prefill
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.12
+    helm.sh/chart: llm-d-modelservice-v0.4.15
     app.kubernetes.io/version: "v0.4.0"
     app.kubernetes.io/managed-by: Helm
 spec: