ComputeCluster is the Schema for the ComputeClusters API. Manages a Machine Learning Compute Cluster.
Type
CRD
Group
machinelearningservices.azure.upbound.io
Version
v1beta1
apiVersion: machinelearningservices.azure.upbound.io/v1beta1
kind: ComputeCluster
ComputeClusterSpec defines the desired state of ComputeCluster
No description provided.
An identity block as defined below. Changing this forces a new Machine Learning Compute Cluster to be created.
Specifies a list of User Assigned Managed Identity IDs to be assigned to this Machine Learning Compute Cluster. Changing this forces a new resource to be created.
Reference to a Workspace in machinelearningservices to populate machineLearningWorkspaceId.
Policies for referencing.
Selector for a Workspace in machinelearningservices to populate machineLearningWorkspaceId.
Policies for selection.
A scale_settings block as defined below. Changing this forces a new Machine Learning Compute Cluster to be created.
Credentials for an administrator user account that will be created on each compute node. A ssh block as defined below. Changing this forces a new Machine Learning Compute Cluster to be created.
Reference to a Subnet in network to populate subnetResourceId.
Policies for referencing.
Selector for a Subnet in network to populate subnetResourceId.
Policies for selection.
THIS IS AN ALPHA FIELD. Do not use it in production. It is not honored unless the relevant Crossplane feature flag is enabled, and may be changed or removed without notice. InitProvider holds the same fields as ForProvider, with the exception of Identifier and other resource reference fields. The fields that are in InitProvider are merged into ForProvider when the resource is created. The same fields are also added to the terraform ignore_changes hook, to avoid updating them after creation. This is useful for fields that are required on creation, but we do not desire to update them after creation, for example because of an external controller is managing them, like an autoscaler.
An identity block as defined below. Changing this forces a new Machine Learning Compute Cluster to be created.
Specifies a list of User Assigned Managed Identity IDs to be assigned to this Machine Learning Compute Cluster. Changing this forces a new resource to be created.
A scale_settings block as defined below. Changing this forces a new Machine Learning Compute Cluster to be created.
Credentials for an administrator user account that will be created on each compute node. A ssh block as defined below. Changing this forces a new Machine Learning Compute Cluster to be created.
THIS IS AN ALPHA FIELD. Do not use it in production. It is not honored unless the relevant Crossplane feature flag is enabled, and may be changed or removed without notice. ManagementPolicies specify the array of actions Crossplane is allowed to take on the managed and external resources. This field is planned to replace the DeletionPolicy field in a future release. Currently, both could be set independently and non-default values would be honored if the feature flag is enabled. If both are custom, the DeletionPolicy field will be ignored. See the design doc for more information: https://github.com/crossplane/crossplane/blob/499895a25d1a1a0ba1604944ef98ac7a1a71f197/design/design-doc-observe-only-resources.md?plain=1#L223 and this one: https://github.com/crossplane/crossplane/blob/444267e84783136daa93568b364a5f01228cacbe/design/one-pager-ignore-changes.md
ProviderConfigReference specifies how the provider that will be used to create, observe, update, and delete this managed resource should be configured.
Policies for referencing.
ProviderReference specifies the provider that will be used to create, observe, update, and delete this managed resource. Deprecated: Please use ProviderConfigReference, i.e. providerConfigRef
Policies for referencing.
PublishConnectionDetailsTo specifies the connection secret config which contains a name, metadata and a reference to secret store config to which any connection details for this managed resource should be written. Connection details frequently include the endpoint, username, and password required to connect to the managed resource.
WriteConnectionSecretToReference specifies the namespace and name of a Secret to which any connection details for this managed resource should be written. Connection details frequently include the endpoint, username, and password required to connect to the managed resource. This field is planned to be replaced in a future release in favor of PublishConnectionDetailsTo. Currently, both could be set independently and connection details would be published to both without affecting each other.
ComputeClusterStatus defines the observed state of ComputeCluster.
No description provided.
An identity block as defined below. Changing this forces a new Machine Learning Compute Cluster to be created.
Specifies a list of User Assigned Managed Identity IDs to be assigned to this Machine Learning Compute Cluster. Changing this forces a new resource to be created.
A scale_settings block as defined below. Changing this forces a new Machine Learning Compute Cluster to be created.
Credentials for an administrator user account that will be created on each compute node. A ssh block as defined below. Changing this forces a new Machine Learning Compute Cluster to be created.
Conditions of the resource.
compcluster001
apiVersion: machinelearningservices.azure.upbound.io/v1beta1
kind: ComputeCluster
metadata:
annotations:
meta.upbound.io/example-id: machinelearningservices/v1beta1/computecluster
uptest.upbound.io/timeout: "2400"
labels:
testing.upbound.io/example-name: compcluster001
name: compcluster001
spec:
forProvider:
location: West Europe
machineLearningWorkspaceIdSelector:
matchLabels:
testing.upbound.io/example-name: example
name: compcluster001
scaleSettings:
- maxNodeCount: 1
minNodeCount: 0
scaleDownNodesAfterIdleDuration: PT10M
subnetResourceIdSelector:
matchLabels:
testing.upbound.io/example-name: example
vmPriority: Dedicated
vmSize: STANDARD_DS11_V2
© 2022 Upbound, Inc.
Discover the building blocksfor your internal cloud platform.