Add training runtimes

Signed-off-by: Brian Gallagher <briangal@gmail.com>

Author: briangallagher

manifests/rhoai/runtimes/kustomization.yaml

@@ -5,3 +5,8 @@ resources:
   - torch_cuda_251.yaml
   - torch_rocm_241.yaml
   - torch_rocm_251.yaml
+  - torch_distributed_th03_cuda128_torch28_py312.yaml
+  - torch_distributed.yaml
+  - training_hub_th03_cuda128_torch28_py312.yaml
+  - training_hub.yaml
+  - training_hub_th03_cuda128_torch28_py312.yaml

manifests/rhoai/runtimes/torch_distributed.yaml

@@ -0,0 +1,26 @@
+apiVersion: trainer.kubeflow.org/v1alpha1
+kind: ClusterTrainingRuntime
+metadata:
+  name: torch-distributed
+  labels:
+    trainer.kubeflow.org/framework: torch
+spec:
+  mlPolicy:
+    numNodes: 1
+    torch:
+      numProcPerNode: auto
+  template:
+    spec:
+      replicatedJobs:
+        - name: node
+          template:
+            metadata:
+              labels:
+                trainer.kubeflow.org/trainjob-ancestor-step: trainer
+            spec:
+              template:
+                spec:
+                  containers:
+                    - name: node
+                      image: quay.io/opendatahub/odh-training-th03-cuda128-torch28-py312-rhel9:latest
+

manifests/rhoai/runtimes/torch_distributed_th03_cuda128_torch28_py312.yaml

@@ -0,0 +1,26 @@
+apiVersion: trainer.kubeflow.org/v1alpha1
+kind: ClusterTrainingRuntime
+metadata:
+  name: torch-distributed
+  labels:
+    trainer.kubeflow.org/framework: torch
+spec:
+  mlPolicy:
+    numNodes: 1
+    torch:
+      numProcPerNode: auto
+  template:
+    spec:
+      replicatedJobs:
+        - name: node
+          template:
+            metadata:
+              labels:
+                trainer.kubeflow.org/trainjob-ancestor-step: trainer
+            spec:
+              template:
+                spec:
+                  containers:
+                    - name: node
+                      image: quay.io/opendatahub/odh-training-th03-cuda128-torch28-py312-rhel9:latest
+

manifests/rhoai/runtimes/training_hub.yaml

@@ -0,0 +1,26 @@
+apiVersion: trainer.kubeflow.org/v1alpha1
+kind: ClusterTrainingRuntime
+metadata:
+  name: training-hub
+  labels:
+    trainer.kubeflow.org/framework: training-hub
+spec:
+  mlPolicy:
+    numNodes: 1
+    torch:
+      numProcPerNode: auto
+  template:
+    spec:
+      replicatedJobs:
+        - name: node
+          template:
+            metadata:
+              labels:
+                trainer.kubeflow.org/trainjob-ancestor-step: trainer
+            spec:
+              template:
+                spec:
+                  containers:
+                    - name: node
+                      image: quay.io/opendatahub/odh-training-th03-cuda128-torch28-py312-rhel9:latest
+

manifests/rhoai/runtimes/training_hub_th03_cuda128_torch28_py312.yaml

@@ -0,0 +1,26 @@
+apiVersion: trainer.kubeflow.org/v1alpha1
+kind: ClusterTrainingRuntime
+metadata:
+  name: training-hub03-cuda128-torch28-py312
+  labels:
+    trainer.kubeflow.org/framework: training-hub
+spec:
+  mlPolicy:
+    numNodes: 1
+    torch:
+      numProcPerNode: auto
+  template:
+    spec:
+      replicatedJobs:
+        - name: node
+          template:
+            metadata:
+              labels:
+                trainer.kubeflow.org/trainjob-ancestor-step: trainer
+            spec:
+              template:
+                spec:
+                  containers:
+                    - name: node
+                      image: quay.io/opendatahub/odh-training-th03-cuda128-torch28-py312-rhel9:latest
+