Add Dynamo support

.github/workflows/release_docker.yaml

@@ -21,10 +21,22 @@ jobs:
           - nominatim
           - postgresql-nominatim
           - cryosparc
+          - vllm-runtime
     name: Release ${{ matrix.image }} Docker image
     steps:
       - name: Checkout
         uses: actions/checkout@v6
+      - name: Free Disk Space (Ubuntu)
+        if: matrix.image == 'vllm-runtime'
+        uses: jlumbroso/free-disk-space@v1.3.1
+        with:
+          tool-cache: false
+          android: true
+          dotnet: true
+          haskell: true
+          large-packages: true
+          docker-images: true
+          swap-storage: true
       - name: Login to DockerHub
         if: matrix.image != 'dcgm-exporter'
         uses: docker/login-action@v4

.github/workflows/test_docker.yaml

@@ -0,0 +1,34 @@
+name: Test Docker
+
+on:
+  pull_request:
+    paths:
+      - docker-images/**
+      - .github/workflows/test_docker.yaml
+
+jobs:
+  test-docker:
+    runs-on: ubuntu-latest
+    strategy:
+      fail-fast: false
+      matrix:
+        image:
+          - vllm-runtime
+    name: Test ${{ matrix.image }} Docker image
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v6
+      - name: Free Disk Space (Ubuntu)
+        if: matrix.image == 'vllm-runtime'
+        uses: jlumbroso/free-disk-space@v1.3.1
+        with:
+          tool-cache: false
+          android: true
+          dotnet: true
+          haskell: true
+          large-packages: true
+          docker-images: true
+          swap-storage: true
+      - name: Build ${{ matrix.image }} image
+        run: |
+          make -f docker-images/${{ matrix.image }}/Makefile build

README.md

@@ -8,6 +8,7 @@ Charts for deploying OSC specific Kubernetes services using Helm
   - [webservice](#webservice)
   - [osc-open-webui](#osc-open-webui)
   - [database](#database)
+  - [dynamo](#dynamo)
 - [Helm Chart Values Patterns](#helm-chart-values-patterns)
   - [Environment-Specific Values](#environment-specific-values)
 - [Infrastructure charts](#infrastructure-charts)
@@ -62,6 +63,10 @@ The [database](charts/database/README.md) chart provides database service deploy
 
 It includes integration with osc-common for consistent security and configuration practices.
 
+### dynamo
+
+The [dynamo](charts/dynamo/README.md) chart handles deploying Dynamo and resources related to Dynamo at OSC.
+
 ## Helm Chart Values Patterns
 
 When using the OSC Puppet infrastructure, some values are automatically added to Helm charts, particularly for webservices.

charts/dynamo/.helmignore

@@ -0,0 +1,23 @@
+# Patterns to ignore when building packages.
+# This supports shell glob matching, relative path matching, and
+# negation (prefixed with !). Only one pattern per line.
+.DS_Store
+# Common VCS dirs
+.git/
+.gitignore
+.bzr/
+.bzrignore
+.hg/
+.hgignore
+.svn/
+# Common backup files
+*.swp
+*.bak
+*.tmp
+*.orig
+*~
+# Various IDEs
+.project
+.idea/
+*.tmproj
+.vscode/

charts/dynamo/Chart.yaml

@@ -0,0 +1,35 @@
+apiVersion: v2
+name: dynamo
+description: A Helm chart to deploy Dynamo resources at OSC
+
+# A chart can be either an 'application' or a 'library' chart.
+#
+# Application charts are a collection of templates that can be packaged into versioned archives
+# to be deployed.
+#
+# Library charts provide useful utilities or functions for the chart developer. They're included as
+# a dependency of application charts to inject those utilities and functions into the rendering
+# pipeline. Library charts do not define any templates and therefore cannot be deployed.
+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: 0.1.0
+
+# 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.
+# It is recommended to use it with quotes.
+appVersion: "1.2.0"
+maintainers:
+  - name: treydock
+dependencies:
+  - name: osc-common
+    version: 0.14.2
+    repository: https://osc.github.io/osc-helm-charts/
+    # repository: file://../osc-common
+  - name: osc-open-webui
+    version: 0.7.3
+    repository: https://osc.github.io/osc-helm-charts/
+    # repository: file://../osc-open-webui

charts/dynamo/README.md

@@ -0,0 +1,98 @@
+# dynamo
+
+![Version: 0.1.0](https://img.shields.io/badge/Version-0.1.0-informational?style=flat-square) ![Type: application](https://img.shields.io/badge/Type-application-informational?style=flat-square) ![AppVersion: 1.2.0](https://img.shields.io/badge/AppVersion-1.2.0-informational?style=flat-square)
+
+A Helm chart to deploy Dynamo resources at OSC
+
+## dynamo Chart
+
+This chart handles the deployment of Dynamo resources.  It depends on the Dynamo platform and operator being deployed.
+
+Components:
+
+* image-puller
+    * A daemonset that pulls the Dynamo images and sleeps to ensure those images are on all nodes
+* Items to support Dynamo
+    * PVC for cache storage
+    * Huggingface secret
+* Dynamo frontend
+* Open WebUI
+
+## Usage
+
+The following is the minimal values to deploy this chart:
+
+```yaml
+---
+global:
+  imagePullSecret:
+    password: <webservices-read password>
+  ingress:
+    host: <Ingress host>
+    hostAlias: <Ingress host alias>
+  webui_secret_key: <Open WebUI webui secret>
+  models:
+    qwen3:
+      model: Qwen/Qwen3-0.6B
+      disagg: true
+      kvRouting: true
+      args:
+        - --max-model-len
+        - "20480"
+      decode:
+        args: []
+        cpu: 2
+        memory: 6Gi
+        gpu: 1
+        gpuType: NVIDIA-A100-PCIE-40GB-MIG-1g.5gb
+        nodes: 1
+      prefill:
+        args: []
+        cpu: 2
+        memory: 6Gi
+        gpu: 1
+        gpuType: NVIDIA-A100-PCIE-40GB-MIG-1g.5gb
+        nodes: 1
+hfToken:
+  value: <Huggingface token>
+```
+
+## Requirements
+
+| Repository | Name | Version |
+|------------|------|---------|
+| https://osc.github.io/osc-helm-charts/ | osc-common | 0.14.2 |
+| https://osc.github.io/osc-helm-charts/ | osc-open-webui | 0.7.3 |
+
+## Values
+
+| Key | Type | Default | Description |
+|-----|------|---------|-------------|
+| global.oscServiceAccount | string | `"dynamo"` | The service account used by OSC deployments. |
+| global.environment | string | `"production"` | The deployment's OSC environment |
+| global.nodeSelectorRole | string | `"webservices"` | The OSC node role to use with nodeSelector |
+| global.imagePullSecret.name | string | `"osc-registry"` | imagePullSecret name |
+| global.imagePullSecret.registry | string | `"docker-registry.osc.edu"` | imagePullSecret registry |
+| global.imagePullSecret.username | string | `"robot$webservices-read"` | imagePullSecret username |
+| global.imagePullSecret.password | string | **required** | imagePullSecret password. This value will be set by OSC's Puppet. |
+| global.fileset | string | `"PZS0645"` | The fileset for storage |
+| global.storageClass | string | `"local-ess"` | The storage class to use for PV claims |
+| global.ingress.host | string | `""` | Ingress host name |
+| global.ingress.hostAlias | string | `""` | Ingress host alias |
+| global.auth.allowGroups | list | `["oscall","PZS0645"]` | Restrict access to these groups |
+| global.webui_secret_key | string | **required** | The Open WebUI secret key |
+| global.models | object | `{}` | Define models |
+| image.repository | string | `"kubernetes/ai-dynamo/vllm-runtime"` | The repository path to main vllm runtime image |
+| image.tag | string | The chart's appVersion | The vllm runtime image tag |
+| image.release | string | `"0"` | The release of the custom OSC vllm runtime image |
+| hfToken.value | string | **required** | The HF token for Hugging Face |
+| cache | object | `{}` |  |
+| defaultGpuType | string | `"NVIDIA-A100-PCIE-40GB-MIG-7g.40gb"` | The default GPU type |
+| rdmaResource | string | `"rdma/shared_mlx5"` | The RDMA resource name in Kubernetes |
+| kvTransferConfig | object | `{"kv_connector":"NixlConnector","kv_role":"kv_both"}` | The configuration of kv-transfer-config |
+| kvEventsConfig | object | `{"enable_kv_cache_events":true,"endpoint":"tcp://*:20080","publisher":"zmq","topic":"kv-events"}` | The configuration for kv-events-config |
+| osc-open-webui.open-webui.image.tag | string | `"0.9.6"` | The version of Open WebUI |
+| osc-open-webui.open-webui.sso.enableRoleManagement | bool | `true` | Enables role access controls in Open WebUI |
+
+----------------------------------------------
+Autogenerated from chart metadata using [helm-docs v1.14.2](https://github.com/norwoodj/helm-docs/releases/v1.14.2)

charts/dynamo/README.md.gotmpl

@@ -0,0 +1,65 @@
+{{ template "chart.header" . }}
+{{ template "chart.deprecationWarning" . }}
+
+{{ template "chart.badgesSection" . }}
+
+{{ template "chart.description" . }}
+
+## dynamo Chart
+
+This chart handles the deployment of Dynamo resources.  It depends on the Dynamo platform and operator being deployed.
+
+Components:
+
+* image-puller
+    * A daemonset that pulls the Dynamo images and sleeps to ensure those images are on all nodes
+* Items to support Dynamo
+    * PVC for cache storage
+    * Huggingface secret
+* Dynamo frontend
+* Open WebUI
+
+## Usage
+
+The following is the minimal values to deploy this chart:
+
+```yaml
+---
+global:
+  imagePullSecret:
+    password: <webservices-read password>
+  ingress:
+    host: <Ingress host>
+    hostAlias: <Ingress host alias>
+  webui_secret_key: <Open WebUI webui secret>
+  models:
+    qwen3:
+      model: Qwen/Qwen3-0.6B
+      disagg: true
+      kvRouting: true
+      args:
+        - --max-model-len
+        - "20480"
+      decode:
+        args: []
+        cpu: 2
+        memory: 6Gi
+        gpu: 1
+        gpuType: NVIDIA-A100-PCIE-40GB-MIG-1g.5gb
+        nodes: 1
+      prefill:
+        args: []
+        cpu: 2
+        memory: 6Gi
+        gpu: 1
+        gpuType: NVIDIA-A100-PCIE-40GB-MIG-1g.5gb
+        nodes: 1
+hfToken:
+  value: <Huggingface token>
+```
+
+{{ template "chart.requirementsSection" . }}
+
+{{ template "chart.valuesSection" . }}
+
+{{ template "helm-docs.versionFooter" . }}

charts/dynamo/templates/_helpers.tpl

@@ -0,0 +1,96 @@
+{{/*
+Expand the name of the chart.
+*/}}
+{{- define "dynamo.name" -}}
+{{- default .Chart.Name .Values.nameOverride | trunc 63 | trimSuffix "-" }}
+{{- end }}
+
+{{/*
+Create a default fully qualified app name.
+We truncate at 63 chars because some Kubernetes name fields are limited to this (by the DNS naming spec).
+If release name contains chart name it will be used as a full name.
+*/}}
+{{- define "dynamo.fullname" -}}
+{{- if .Values.fullnameOverride }}
+{{- .Values.fullnameOverride | trunc 63 | trimSuffix "-" }}
+{{- else }}
+{{- $name := default .Chart.Name .Values.nameOverride }}
+{{- if contains $name .Release.Name }}
+{{- .Release.Name | trunc 63 | trimSuffix "-" }}
+{{- else }}
+{{- printf "%s-%s" .Release.Name $name | trunc 63 | trimSuffix "-" }}
+{{- end }}
+{{- end }}
+{{- end }}
+
+{{/*
+Create chart name and version as used by the chart label.
+*/}}
+{{- define "dynamo.chart" -}}
+{{- printf "%s-%s" .Chart.Name .Chart.Version | replace "+" "_" | trunc 63 | trimSuffix "-" }}
+{{- end }}
+
+{{/*
+Common labels
+*/}}
+{{- define "dynamo.labels" -}}
+helm.sh/chart: {{ include "dynamo.chart" . }}
+{{ include "dynamo.selectorLabels" . }}
+{{- if .Chart.AppVersion }}
+app.kubernetes.io/version: {{ .Chart.AppVersion | quote }}
+{{- end }}
+app.kubernetes.io/managed-by: {{ .Release.Service }}
+{{- end }}
+
+{{/*
+Selector labels
+*/}}
+{{- define "dynamo.selectorLabels" -}}
+app.kubernetes.io/name: {{ include "dynamo.name" . }}
+app.kubernetes.io/instance: {{ .Release.Name }}
+{{- end }}
+
+{{/*
+Create the name of the service account to use
+*/}}
+{{- define "dynamo.serviceAccountName" -}}
+{{- if .Values.serviceAccount.create }}
+{{- default (include "dynamo.fullname" .) .Values.serviceAccount.name }}
+{{- else }}
+{{- default "default" .Values.serviceAccount.name }}
+{{- end }}
+{{- end }}
+
+{{/*
+image-puller labels
+*/}}
+{{- define "dynamo.image-puller.labels" -}}
+helm.sh/chart: {{ include "dynamo.chart" . }}
+{{ include "dynamo.image-puller.selectorLabels" . }}
+{{- if .Chart.AppVersion }}
+app.kubernetes.io/version: {{ .Chart.AppVersion | quote }}
+{{- end }}
+app.kubernetes.io/managed-by: {{ .Release.Service }}
+{{- end }}
+
+{{/*
+image-puller selector labels
+*/}}
+{{- define "dynamo.image-puller.selectorLabels" -}}
+app.kubernetes.io/name: image-puller
+app.kubernetes.io/instance: {{ .Release.Name }}
+{{- end }}
+
+{{/* Image tag helper */}}
+{{- define "dynamo.image" -}}
+{{- $tag := printf "%s-osc-r%s" (.Values.image.tag | default .Chart.AppVersion) .Values.image.release }}
+{{- printf "%s/%s:%s" .Values.global.imagePullSecret.registry .Values.image.repository $tag }}
+{{- end }}
+
+{{- define "dynamo.openai.urls" -}}
+{{- $urls := list }}
+{{- range $name, $model := .Values.global.models }}
+    {{- $urls = append $urls (printf "http://%s-frontend.%s.svc.cluster.local:8000/v1" $name $.Release.Namespace) }}
+{{- end }}
+{{- $urls | join ";" }}
+{{- end }}