Merge pull request #316 from codekow/rhoai-update

update: rhoai

.wordlist-md

@@ -17,12 +17,17 @@ CICD
 CRD
 CRDs
 CheCluster
+ClusterPolicy
 ClusterRoleBindings
 ClusterTask
+CodeFlare
 CodeReady
 Codeflare
+ConfigMap
+DCGM
 DNS
 DataScienceCluster
+DataSciencePipelines
 Dev
 DevSpaces
 DevWorkspace
@@ -43,6 +48,7 @@ IgnoreExtraneous
 JSON
 Jaeger
 Jenkins
+KServe
 Kafka
 Keycloak
 Kibana
@@ -59,6 +65,7 @@ MTA
 MachineConfig
 MachineSet
 Minio
+ModelMesh
 MultiClusterHub
 NFD
 NMState
@@ -80,6 +87,7 @@ Querier
 RBAC
 README
 RHACM
+RHOAI
 Rollout
 SAR
 SDKs
@@ -118,6 +126,7 @@ approver
 argocd
 argoproj
 arn
+auth
 authorino
 autoscaling
 aws
@@ -129,11 +138,14 @@ checluster
 cicd
 cli
 clusterissuer
+clusterpolicies
 clusterrolebinding
 composable
 config
 configmap
+culler
 datasource
+dcgmExporter
 deployable
 dev
 devfile
@@ -171,10 +183,14 @@ io
 jaeger
 jenkins
 json
+jsonpath
 kafka
 kataconfig
 kiali
 knative
+kubeadmin
+kubectl
+kubernetes
 kustomization
 kustomize
 labeler
@@ -211,10 +227,13 @@ postgres
 pre
 prem
 prometheus
+pvc
 redhat
 redistributions
 repo
+rh
 rhel
+rhoai
 rhpds
 runtime
 sagemaker
@@ -233,6 +252,7 @@ tekton
 templating
 thanos
 traefik
+triton
 truly
 vSphere
 vscode

openshift-ai

@@ -0,0 +1 @@
+rhods-operator

rhods-operator/INFO.md

@@ -0,0 +1,22 @@
+# rhods-operator
+
+Red Hat OpenShift AI is a complete platform for the entire lifecycle of your AI/ML projects.
+
+When using Red Hat OpenShift AI, your users will find all the tools they would expect from a modern AI/ML platform in an interface that is intuitive, requires no local install, and is backed by the power of your OpenShift cluster.
+
+Your Data Scientists will feel right at home with quick and simple access to the Notebook interface they are used to. They can leverage the default Notebook Images (Including PyTorch, tensorflow, and CUDA), or add custom ones. Your MLOps engineers will be able to leverage Data Science Pipelines to easily parallelize and/or schedule the required workloads. They can then quickly serve, monitor, and update the created AI/ML models. They can do that by either using the provided out-of-the-box OpenVino Server Model Runtime or by adding their own custom serving runtime instead. These activities are tied together with the concept of Data Science Projects, simplifying both organization and collaboration.
+
+But beyond the individual features, one of the key aspects of this platform is its flexibility. Not only can you augment it with your own Customer Workbench Image and Custom Model Serving Runtime Images, but you will also have a consistent experience across any infrastructure footprint. Be it in the public cloud, private cloud, on-premises, and even in disconnected clusters. Red Hat OpenShift AI can be installed on any supported OpenShift. It can scale out or in depending on the size of your team and its computing requirements.
+
+Finally, thanks to the operator-driven deployment and updates, the administrative load of the platform is very light, leaving everyone more time to focus on the work that makes a difference.
+
+### Components
+* Dashboard
+* Curated Workbench Images (incl CUDA, PyTorch, Tensorflow, code-server)
+* Ability to add Custom Images
+* Ability to leverage accelerators (such as NVIDIA GPU)
+* Data Science Pipelines (including Elyra notebook interface)
+* Model Serving using ModelMesh and Kserve.
+* Ability to use other runtimes for serving
+* Model Monitoring
+* Distributed workloads (KubeRay, CodeFlare, Kueue)

openshift-ai/README.md → rhods-operator/README.md

@@ -6,23 +6,26 @@ Do not use the `base` directory directly, as you will need to patch the `channel
 
 The current *overlays* available are for the following channels:
 
+* [alpha](operator/overlays/alpha)
 * [beta](operator/overlays/beta)
-* [fast](operator/overlays/beta)
+* [embedded](operator/overlays/embedded)
+* [eus-2.8](operator/overlays/eus-2.8)
+* [fast](operator/overlays/fast)
 * [stable](operator/overlays/stable)
-
+* [stable-2.8](operator/overlays/stable-2.8)
 
 ## Usage
 
 If you have cloned the `gitops-catalog` repository, you can install Red Hat OpenShift AI based on the overlay of your choice by running from the root (`gitops-catalog`) directory.
 
 ```
-oc apply -k openshift-ai/operator/overlays/<channel>
+oc apply -k rhods-operator/operator/overlays/<channel>
 ```
 
 Or, without cloning:
 
 ```
-oc apply -k https://github.com/redhat-cop/gitops-catalog/openshift-ai/operator/overlays/<channel>
+oc apply -k https://github.com/redhat-cop/gitops-catalog/rhods-operator/operator/overlays/<channel>
 ```
 
 As part of a different overlay in your own GitOps repo:
@@ -31,5 +34,5 @@ As part of a different overlay in your own GitOps repo:
 apiVersion: kustomize.config.k8s.io/v1beta1
 kind: Kustomization
 resources:
-  - https://github.com/redhat-cop/gitops-catalog/openshift-ai/operator/overlays/<channel>?ref=main
+  - https://github.com/redhat-cop/gitops-catalog/rhods-operator/operator/overlays/<channel>?ref=main
 ```

rhods-operator/aggregate/overlays/eus-2.8-nvidia-gpu/kustomization.yaml

@@ -0,0 +1,9 @@
+apiVersion: kustomize.config.k8s.io/v1beta1
+kind: Kustomization
+
+commonAnnotations:
+  argocd.argoproj.io/sync-options: SkipDryRunOnMissingResource=true
+
+resources:
+  - ../../../operator/overlays/eus-2.8
+  - ../../../instance/overlays/eus-2.8-nvidia-gpu

rhods-operator/aggregate/overlays/eus-2.8/kustomization.yaml

@@ -0,0 +1,9 @@
+apiVersion: kustomize.config.k8s.io/v1beta1
+kind: Kustomization
+
+commonAnnotations:
+  argocd.argoproj.io/sync-options: SkipDryRunOnMissingResource=true
+
+resources:
+  - ../../../operator/overlays/eus-2.8
+  - ../../../instance/overlays/eus-2.8

rhods-operator/aggregate/overlays/fast-nvidia-gpu/kustomization.yaml

@@ -0,0 +1,9 @@
+apiVersion: kustomize.config.k8s.io/v1beta1
+kind: Kustomization
+
+commonAnnotations:
+  argocd.argoproj.io/sync-options: SkipDryRunOnMissingResource=true
+
+resources:
+  - ../../../operator/overlays/fast
+  - ../../../instance/overlays/fast-nvidia-gpu

openshift-ai/aggregate/overlays/default/kustomization.yaml → rhods-operator/aggregate/overlays/fast/kustomization.yaml

@@ -5,5 +5,5 @@ commonAnnotations:
   argocd.argoproj.io/sync-options: SkipDryRunOnMissingResource=true
 
 resources:
-  - ../../../operator/overlays/stable
-  - ../../../instance/overlays/default
+  - ../../../operator/overlays/fast
+  - ../../../instance/overlays/fast

openshift-ai/instance/base/datasciencecluster.yaml → rhods-operator/instance/base/datasciencecluster.yaml

@@ -6,21 +6,25 @@ spec:
   components:
     codeflare:
       managementState: Removed
-    dashboard:
-      managementState: Managed
-    datasciencepipelines:
-      managementState: Managed
     kserve:
-      managementState: Managed
+      managementState: Removed
       serving:
         ingressGateway:
           certificate:
             type: SelfSigned
         managementState: Managed
         name: knative-serving
-    modelmeshserving:
-      managementState: Managed
+    trustyai:
+      managementState: Removed
     ray:
       managementState: Removed
+    kueue:
+      managementState: Removed
     workbenches:
+      managementState: Removed
+    dashboard:
       managementState: Managed
+    modelmeshserving:
+      managementState: Removed
+    datasciencepipelines:
+      managementState: Removed

rhods-operator/instance/base/dsc-init.yaml

@@ -0,0 +1,18 @@
+apiVersion: dscinitialization.opendatahub.io/v1
+kind: DSCInitialization
+metadata:
+  name: default-dsci
+spec:
+  applicationsNamespace: redhat-ods-applications
+  monitoring:
+    managementState: Managed
+    namespace: redhat-ods-monitoring
+  serviceMesh:
+    controlPlane:
+      metricsCollection: Istio
+      name: data-science-smcp
+      namespace: istio-system
+    managementState: Removed
+  trustedCABundle:
+    customCABundle: ''
+    managementState: Managed

openshift-ai/instance/base/kustomization.yaml → rhods-operator/instance/base/kustomization.yaml

@@ -1,5 +1,10 @@
 apiVersion: kustomize.config.k8s.io/v1beta1
 kind: Kustomization
 
+namespace: redhat-ods-applications
+
 resources:
   - datasciencecluster.yaml
+  - dsc-init.yaml
+  - namespace.yaml
+  - odhdashboardconfig.yaml

rhods-operator/instance/base/namespace.yaml

@@ -0,0 +1,4 @@
+apiVersion: v1
+kind: Namespace
+metadata:
+  name: redhat-ods-applications

rhods-operator/instance/base/odhdashboardconfig.yaml

@@ -0,0 +1,6 @@
+apiVersion: opendatahub.io/v1alpha
+kind: OdhDashboardConfig
+metadata:
+  annotations:
+    argocd.argoproj.io/sync-options: ServerSideApply=true
+  name: odh-dashboard-config

rhods-operator/instance/components/auth-with-authorino/README.md

@@ -0,0 +1,21 @@
+# components-distributed-compute
+
+## Purpose
+This component is designed help configure the the authentication endpoint for RHOAI with KServe using Authorino with ServiceMesh
+
+The Authorino auth is available in RHOAI 2.9 and later.
+
+## Usage
+
+This component can be added to a base by adding the `components` section to your overlay `kustomization.yaml` file:
+
+```
+apiVersion: kustomize.config.k8s.io/v1beta1
+kind: Kustomization
+
+resources:
+  - ../../base
+
+components:
+  - ../../components/auth-with-authorino
+```

rhods-operator/instance/components/auth-with-authorino/kustomization.yaml

@@ -0,0 +1,7 @@
+apiVersion: kustomize.config.k8s.io/v1alpha1
+kind: Component
+
+patches:
+  - path: patch-dsc-init.yaml
+    target:
+      kind: DSCInitialization

rhods-operator/instance/components/auth-with-authorino/patch-dsc-init.yaml

@@ -0,0 +1,9 @@
+apiVersion: dscinitialization.opendatahub.io/v1
+kind: DSCInitialization
+metadata:
+  name: default-dsci
+spec:
+  serviceMesh:
+    auth:
+      audiences:
+        - 'https://kubernetes.default.svc'

rhods-operator/instance/components/components-distributed-compute/README.md

@@ -0,0 +1,24 @@
+# components-distributed-compute
+
+## Purpose
+This component is designed help configure the distributed compute specific components including the following items:
+
+CodeFlare
+Ray
+
+The Distributed Compute Components are Generally Available as of RHOAI 2.9.
+
+## Usage
+
+This component can be added to a base by adding the `components` section to your overlay `kustomization.yaml` file:
+
+```
+apiVersion: kustomize.config.k8s.io/v1beta1
+kind: Kustomization
+
+resources:
+  - ../../base
+
+components:
+  - ../../components/components-distributed-compute
+```

rhods-operator/instance/components/components-distributed-compute/kustomization.yaml

@@ -0,0 +1,7 @@
+apiVersion: kustomize.config.k8s.io/v1alpha1
+kind: Component
+
+patches:
+  - path: patch-datasciencecluster.yaml
+    target:
+      kind: DataScienceCluster

rhods-operator/instance/components/components-distributed-compute/patch-datasciencecluster.yaml

@@ -0,0 +1,12 @@
+kind: DataScienceCluster
+apiVersion: datasciencecluster.opendatahub.io/v1
+metadata:
+  name: default
+spec:
+  components:
+    codeflare:
+      managementState: Managed
+    kueue:
+      managementState: Managed
+    ray:
+      managementState: Managed