Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

update: OpenShift AI #342

Merged
merged 2 commits into from
Dec 9, 2024
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 6 additions & 0 deletions .wordlist-md
Original file line number Diff line number Diff line change
Expand Up @@ -59,6 +59,7 @@ KieApp
Koku
KokuMetricsConfig
Kubernetes
Kueue
Kyverno
LDAP
LVM
Expand Down Expand Up @@ -116,9 +117,11 @@ Tekton
TempoMonolithic
TempoStack
Thanos
TrustyAI
UI
VolSync
Workspaces
XAI
Zipkin
ack
acm
Expand Down Expand Up @@ -206,6 +209,7 @@ linux
machineset
microservices
mig
modelregistry
namespace
namespaced
namespaces
Expand Down Expand Up @@ -241,6 +245,7 @@ rh
rhel
rhoai
rhpds
rolebinding
runtime
sagemaker
sandboxed
Expand All @@ -261,6 +266,7 @@ thanos
traefik
triton
truly
trustyai
vSphere
vscode
vsphere
Expand Down
6 changes: 4 additions & 2 deletions rhods-operator/INFO.md
Original file line number Diff line number Diff line change
Expand Up @@ -12,11 +12,13 @@ Finally, thanks to the operator-driven deployment and updates, the administrativ

### Components
* Dashboard
* Curated Workbench Images (incl CUDA, PyTorch, Tensorflow, code-server)
* Curated Workbench Images (including CUDA, PyTorch, TensorFlow, code-server, TrustyAI)
* 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)
* Distributed workloads (KubeRay, CodeFlare, Kueue, Training Operator)
* XAI explanations of predictive models (TrustyAI)
* Index and manage models, versions, and artifacts metadata (Model Registry)
3 changes: 3 additions & 0 deletions rhods-operator/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -10,9 +10,12 @@ The current *overlays* available are for the following channels:
* [beta](operator/overlays/beta)
* [embedded](operator/overlays/embedded)
* [eus-2.8](operator/overlays/eus-2.8)
* [eus-2.16](operator/overlays/eus-2.16)
strangiato marked this conversation as resolved.
Show resolved Hide resolved
* [fast](operator/overlays/fast)
* [stable](operator/overlays/stable)
* [stable-2.8](operator/overlays/stable-2.8)
* [stable-2.13](operator/overlays/stable-2.13)
* [stable-2.16](operator/overlays/stable-2.16)

## Usage

Expand Down
Original file line number Diff line number Diff line change
@@ -0,0 +1,9 @@
apiVersion: kustomize.config.k8s.io/v1beta1
kind: Kustomization

commonAnnotations:
argocd.argoproj.io/sync-options: SkipDryRunOnMissingResource=true

resources:
- ../../../operator/overlays/eus-2.16
- ../../../instance/overlays/eus-2.16
Original file line number Diff line number Diff line change
@@ -1,4 +1,4 @@
# components-distributed-compute
# components-auth-with-authorino

## Purpose
This component is designed help configure the the authentication endpoint for RHOAI with KServe using Authorino with ServiceMesh
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -5,6 +5,7 @@ This component is designed help configure the distributed compute specific compo

CodeFlare
Ray
Kueue

The Distributed Compute Components are Generally Available as of RHOAI 2.9.

Expand Down
Original file line number Diff line number Diff line change
@@ -0,0 +1,19 @@
# components-modelregistry

## Purpose
This component is designed help index and manage models, versions, and artifacts metadata

## 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-modelregistry
```
Original file line number Diff line number Diff line change
@@ -0,0 +1,7 @@
apiVersion: kustomize.config.k8s.io/v1alpha1
kind: Component

patches:
- path: patch-datasciencecluster.yaml
target:
kind: DataScienceCluster
Original file line number Diff line number Diff line change
@@ -0,0 +1,9 @@
kind: DataScienceCluster
apiVersion: datasciencecluster.opendatahub.io/v1
metadata:
name: default
spec:
components:
modelregistry:
managementState: Managed
registriesNamespace: rhoai-model-registries
Original file line number Diff line number Diff line change
Expand Up @@ -20,3 +20,5 @@ resources:
components:
- ../../components/components-serving
```

You can customize the PVC size by updating the [patch-rhoai-dashboard.yaml](./patch-rhoai-dashboard.yaml) file.
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,7 @@ spec:
components:
kserve:
managementState: Managed
defaultDeploymentMode: Serverless
strangiato marked this conversation as resolved.
Show resolved Hide resolved
serving:
ingressGateway:
certificate:
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -19,4 +19,4 @@ resources:

components:
- ../../components/components-training
```
```
Original file line number Diff line number Diff line change
@@ -0,0 +1,20 @@
# components-trustyai

## Purpose
This component is designed help XAI explanations of predictive models.
The TrustyAI Components are Generally Available as of RHOAI 2.15.

## 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-trustyai
```
Original file line number Diff line number Diff line change
@@ -0,0 +1,7 @@
apiVersion: kustomize.config.k8s.io/v1alpha1
kind: Component

patches:
- path: patch-datasciencecluster.yaml
target:
kind: DataScienceCluster
Original file line number Diff line number Diff line change
@@ -0,0 +1,8 @@
kind: DataScienceCluster
apiVersion: datasciencecluster.opendatahub.io/v1
metadata:
name: default
spec:
components:
trustyai:
managementState: Managed
Original file line number Diff line number Diff line change
Expand Up @@ -17,7 +17,5 @@ resources:
- ../../base

components:
- ../../components/idle-notebook-culling
```

You can customize the time notebooks can remain running while inactive by updating the [culler-config.yaml](./culler-config.yaml) file.
- ../../components/rolebinding.yaml
```
Original file line number Diff line number Diff line change
Expand Up @@ -19,5 +19,3 @@ resources:
components:
- ../../components/nvidia-gpu-accelerator-profile
```

You can customize the pod sizes by updating the [patch-rhoai-dashboard.yaml](./patch-rhoai-dashboard.yaml) file.
Original file line number Diff line number Diff line change
Expand Up @@ -4,12 +4,9 @@ set -e
TIMEOUT_SECONDS=60

wait_for_service_mesh(){
echo "Checking status of all service_mesh pre-reqs"
echo "Checking status of all service_mesh and serverless and serverless pre-reqs"
SERVICEMESH_RESOURCES=(
crd/kialis.kiali.io:condition=established \
crd/jaegers.jaegertracing.io:condition=established \
crd/elasticsearches.logging.openshift.io:condition=established \
crd/knativeservings.operator.knative.dev:condition=established \
crd/knativeservings.operator.knative.dev:condition=established
crd/servicemeshcontrolplanes.maistra.io:condition=established \
)

Expand Down
Original file line number Diff line number Diff line change
@@ -0,0 +1,10 @@
apiVersion: kustomize.config.k8s.io/v1beta1
kind: Kustomization

resources:
- ../../base

components:
- ../../components/wait-for-servicemesh
- ../../components/components-modelregistry
- ../../components/rhoai-dashboard-access
Original file line number Diff line number Diff line change
@@ -0,0 +1,20 @@
apiVersion: kustomize.config.k8s.io/v1beta1
kind: Kustomization

namespace: redhat-ods-applications

resources:
- ../../base

components:
- ../../components/components-serving
- ../../components/components-training
- ../../components/components-trustyai
- ../../components/default-notebook-pvc-size
- ../../components/idle-notebook-culling
- ../../components/notebook-pod-sizes
- ../../components/nvidia-gpu-accelerator-profile
- ../../components/make-kubeadmin-cluster-admin
- ../../components/model-server-pod-sizes
- ../../components/rhoai-dashboard-access
- ../../components/auth-with-authorino
25 changes: 25 additions & 0 deletions rhods-operator/instance/overlays/eus-2.16/kustomization.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,25 @@
apiVersion: kustomize.config.k8s.io/v1beta1
kind: Kustomization

namespace: redhat-ods-applications

resources:
- ../../base

components:
- ../../components/components-serving
- ../../components/components-training
- ../../components/components-trustyai
- ../../components/default-notebook-pvc-size
- ../../components/idle-notebook-culling
- ../../components/notebook-pod-sizes
- ../../components/make-kubeadmin-cluster-admin
- ../../components/model-server-pod-sizes
- ../../components/rhoai-dashboard-access
- ../../components/auth-with-authorino

patches:
- target:
kind: DataScienceCluster
name: default
path: patch-cert.yaml
3 changes: 3 additions & 0 deletions rhods-operator/instance/overlays/eus-2.16/patch-cert.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,3 @@
- op: replace
path: /spec/components/kserve/serving/ingressGateway/certificate/type
value: OpenshiftDefaultIngress
Original file line number Diff line number Diff line change
Expand Up @@ -9,6 +9,7 @@ components:
- ../../components/components-distributed-compute
- ../../components/components-serving
- ../../components/components-training
- ../../components/components-trustyai
- ../../components/default-notebook-pvc-size
- ../../components/idle-notebook-culling
- ../../components/notebook-pod-sizes
Expand Down
11 changes: 11 additions & 0 deletions rhods-operator/operator/overlays/eus-2.16/kustomization.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,11 @@
apiVersion: kustomize.config.k8s.io/v1beta1
kind: Kustomization

resources:
- ../../base

patches:
- target:
kind: Subscription
name: rhods-operator
path: patch-channel.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,3 @@
- op: replace
path: /spec/channel
value: eus-2.16
11 changes: 11 additions & 0 deletions rhods-operator/operator/overlays/stable-2.13/kustomization.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,11 @@
apiVersion: kustomize.config.k8s.io/v1beta1
kind: Kustomization

resources:
- ../../base

patches:
- target:
kind: Subscription
name: rhods-operator
path: patch-channel.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,3 @@
- op: replace
path: /spec/channel
value: stable-2.13
11 changes: 11 additions & 0 deletions rhods-operator/operator/overlays/stable-2.16/kustomization.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,11 @@
apiVersion: kustomize.config.k8s.io/v1beta1
kind: Kustomization

resources:
- ../../base

patches:
- target:
kind: Subscription
name: rhods-operator
path: patch-channel.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,3 @@
- op: replace
path: /spec/channel
value: stable-2.16
Loading