Skip to content

[Inference Operator] Update CRDs and Helm chart to v3.1 #412

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.

Sign up for GitHub

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

Open
zicanl-amazon wants to merge 1 commit into
base: main
Choose a base branch
from
Open

[Inference Operator] Update CRDs and Helm chart to v3.1 #412

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
4 changes: 2 additions & 2 deletions helm_chart/HyperPodHelmChart/charts/inference-operator/Chart.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,7 +15,7 @@ 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: 2.1.0
version: 2.2.0

# This is the version number of the application being deployed. Keep this aligned
# with operator image MAJOR.MINOR version.
Expand Down Expand Up @@ -50,4 +50,4 @@ dependencies:
alias: keda
version: 2.17.1
repository: "https://kedacore.github.io/charts"
condition: keda.enabled
condition: keda.enabled
307 changes: 302 additions & 5 deletions ...ence-operator/config/crd/inference.sagemaker.aws.amazon.com_inferenceendpointconfigs.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -270,7 +270,7 @@ spec:
type: string
serverAddress:
description: Server address for AMP workspace
pattern: ^https:\/\/aps-workspaces\.[a-zA-Z0-9-]+(?:\.[a-zA-Z0-9-]+)*\.amazonaws\.com\/workspaces\/ws-[a-zA-Z0-9-]+\/[a-zA-Z0-9-]+$|^$
pattern: ^https:\/\/aps-workspaces\.[a-zA-Z0-9-]+(?:\.[a-zA-Z0-9-]+)*\.amazonaws\.com\/workspaces\/ws-[a-zA-Z0-9-]+$|^$
type: string
targetValue:
description: Target metric value for scaling
Expand Down Expand Up @@ -315,7 +315,7 @@ spec:
type: string
serverAddress:
description: Server address for AMP workspace
pattern: ^https:\/\/aps-workspaces\.[a-zA-Z0-9-]+(?:\.[a-zA-Z0-9-]+)*\.amazonaws\.com\/workspaces\/ws-[a-zA-Z0-9-]+\/[a-zA-Z0-9-]+$|^$
pattern: ^https:\/\/aps-workspaces\.[a-zA-Z0-9-]+(?:\.[a-zA-Z0-9-]+)*\.amazonaws\.com\/workspaces\/ws-[a-zA-Z0-9-]+$|^$
type: string
targetValue:
description: Target metric value for scaling
Expand Down Expand Up @@ -343,6 +343,211 @@ spec:
minimum: 0
type: integer
type: object
dataCapture:
description: Configuration for data capture across multiple tiers
(SageMaker, LoadBalancer, Model Pod)
properties:
loadBalancer:
description: Configuration for LoadBalancer level data capture
(Tier 2)
properties:
enabled:
description: Enable or disable load balancer access logs
type: boolean
required:
- enabled
type: object
modelPod:
description: Configuration for Model Pod level data capture (Tier
3)
properties:
bufferConfig:
description: Configuration for buffering and flushing captured
data
properties:
batchSize:
default: 10
description: Number of records to batch before writing
to S3
format: int32
maximum: 1000
minimum: 1
type: integer
flushIntervalSeconds:
default: 60
description: Flush interval in seconds
format: int32
maximum: 300
minimum: 10
type: integer
type: object
captureContentTypeHeader:
description: Configuration for how to treat different content
type headers during capture
properties:
csvContentTypes:
description: |-
List of content type headers to treat as CSV
Each item must be 1-256 characters and match pattern: [a-zA-Z0-9](-*[a-zA-Z0-9])*/[a-zA-Z0-9](-*[a-zA-Z0-9.])*
Example: text/csv, application/csv
items:
type: string
maxItems: 10
minItems: 1
type: array
x-kubernetes-list-type: set
jsonContentTypes:
description: |-
List of content type headers to treat as JSON
Each item must be 1-256 characters and match pattern: [a-zA-Z0-9](-*[a-zA-Z0-9])*/[a-zA-Z0-9](-*[a-zA-Z0-9.])*
Example: application/json, application/jsonlines
items:
type: string
maxItems: 10
minItems: 1
type: array
x-kubernetes-list-type: set
type: object
captureOptions:
description: Capture options (Input, Output, or both). Defaults
to [Input, Output] when enabled.
items:
description: CaptureOption defines what data to capture
(input, output, or both).
properties:
captureMode:
description: 'Capture mode: Input or Output'
enum:
- Input
- Output
type: string
required:
- captureMode
type: object
maxItems: 32
minItems: 1
type: array
enabled:
description: Enable or disable model pod data capture
type: boolean
initialSamplingPercentage:
description: Percentage of requests to capture (0-100). Defaults
to 100 when enabled.
format: int32
maximum: 100
minimum: 0
type: integer
kmsKeyId:
description: Optional KMS key ID, ARN, alias name, or alias
ARN for encrypting captured data
maxLength: 2048
pattern: ^[a-zA-Z0-9:/_-]*$
type: string
payloadConfig:
description: Configuration for payload size limits
properties:
maxPayloadSizeKB:
default: 0
description: Maximum payload size in KB to capture. 0
means no limit (capture full payload).
format: int32
maximum: 10240
minimum: 0
type: integer
type: object
required:
- enabled
type: object
s3Uri:
description: |-
Common S3 URI for all data capture tiers. Each tier will write to a specific prefix within this bucket.
Must use s3:// protocol (required by ALB access logs).
If not provided, the TLS certificate bucket will be used for data capture storage.
maxLength: 512
pattern: ^s3://([^/]+)(/[^,=]*)?$
type: string
sagemakerEndpoint:
description: Configuration for SageMaker Endpoint level data capture
(Tier 1)
properties:
captureContentTypeHeader:
description: Configuration for how to treat different content
type headers during capture
properties:
csvContentTypes:
description: |-
List of content type headers to treat as CSV
Each item must be 1-256 characters and match pattern: [a-zA-Z0-9](-*[a-zA-Z0-9])*/[a-zA-Z0-9](-*[a-zA-Z0-9.])*
Example: text/csv, application/csv
items:
type: string
maxItems: 10
minItems: 1
type: array
x-kubernetes-list-type: set
jsonContentTypes:
description: |-
List of content type headers to treat as JSON
Each item must be 1-256 characters and match pattern: [a-zA-Z0-9](-*[a-zA-Z0-9])*/[a-zA-Z0-9](-*[a-zA-Z0-9.])*
Example: application/json, application/jsonlines
items:
type: string
maxItems: 10
minItems: 1
type: array
x-kubernetes-list-type: set
type: object
captureOptions:
description: Capture options (Input, Output, or both). Defaults
to [Input, Output] when enabled.
items:
description: CaptureOption defines what data to capture
(input, output, or both).
properties:
captureMode:
description: 'Capture mode: Input or Output'
enum:
- Input
- Output
type: string
required:
- captureMode
type: object
maxItems: 32
minItems: 1
type: array
enabled:
description: Enable or disable SageMaker endpoint data capture
type: boolean
initialSamplingPercentage:
description: Percentage of requests to capture (0-100). Defaults
to 100 when enabled.
format: int32
maximum: 100
minimum: 0
type: integer
kmsKeyId:
description: Optional KMS key ID, ARN, alias name, or alias
ARN for encrypting captured data
maxLength: 2048
pattern: ^[a-zA-Z0-9:/_-]*$
type: string
required:
- enabled
type: object
type: object
dnsConfig:
description: DNS automation configuration for Route53. Requires tlsConfig.customCertificateConfig
to be set.
properties:
hostedZoneId:
description: Route53 Hosted Zone ID where the DNS record will
be created.
pattern: ^Z[A-Z0-9]+$
type: string
required:
- hostedZoneId
type: object
endpointName:
description: |-
Name used for Sagemaker Endpoint
Expand Down Expand Up @@ -589,7 +794,7 @@ spec:
type: string
serverAddress:
description: Server address for AMP workspace
pattern: ^https:\/\/aps-workspaces\.[a-zA-Z0-9-]+(?:\.[a-zA-Z0-9-]+)*\.amazonaws\.com\/workspaces\/ws-[a-zA-Z0-9-]+\/[a-zA-Z0-9-]+$|^$
pattern: ^https:\/\/aps-workspaces\.[a-zA-Z0-9-]+(?:\.[a-zA-Z0-9-]+)*\.amazonaws\.com\/workspaces\/ws-[a-zA-Z0-9-]+$|^$
type: string
targetValue:
description: Target metric value for scaling
Expand Down Expand Up @@ -636,7 +841,7 @@ spec:
type: string
serverAddress:
description: Server address for AMP workspace
pattern: ^https:\/\/aps-workspaces\.[a-zA-Z0-9-]+(?:\.[a-zA-Z0-9-]+)*\.amazonaws\.com\/workspaces\/ws-[a-zA-Z0-9-]+\/[a-zA-Z0-9-]+$|^$
pattern: ^https:\/\/aps-workspaces\.[a-zA-Z0-9-]+(?:\.[a-zA-Z0-9-]+)*\.amazonaws\.com\/workspaces\/ws-[a-zA-Z0-9-]+$|^$
type: string
targetValue:
description: Target metric value for scaling
Expand Down Expand Up @@ -2230,6 +2435,14 @@ spec:
maxLength: 253
pattern: ^[a-z0-9]([-a-z0-9]*[a-z0-9])?(\.[a-z0-9]([-a-z0-9]*[a-z0-9])?)*$
type: string
serviceAccountName:
description: |-
Name of the Kubernetes ServiceAccount to use for the inference pod.
If not specified, the namespace's default service account will be used.
This is useful for providing AWS credentials via IRSA to init containers or the worker.
maxLength: 253
pattern: ^[a-z0-9]([-a-z0-9]*[a-z0-9])?(\.[a-z0-9]([-a-z0-9]*[a-z0-9])?)*$
type: string
volumes:
description: |-
Additional volumes to add to the pod spec.
Expand Down Expand Up @@ -4249,13 +4462,59 @@ spec:
required:
- fileSystemId
type: object
huggingFaceModel:
description: HuggingFace model configuration. Required when modelSourceType
is "huggingface".
properties:
commitSHA:
description: |-
Git commit SHA for the model revision. Must be a full 40-character lowercase hex SHA.
If not provided, the operator defaults to "main" branch.
pattern: ^[0-9a-f]{40}$
type: string
modelId:
description: HuggingFace Hub model identifier in org/model
format (e.g. "meta-llama/Llama-3.1-8B-Instruct").
pattern: ^[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+$
type: string
tokenSecretRef:
description: |-
Reference to a Kubernetes Secret containing the HuggingFace API token.
The token is injected as the HF_TOKEN environment variable into the InitContainer only.
properties:
key:
description: The key of the secret to select from. Must
be a valid secret key.
type: string
name:
default: ""
description: |-
Name of the referent.
This field is effectively required, but due to backwards compatibility is
allowed to be empty. Instances of this type with an empty value here are
almost certainly wrong.
More info: https://kubernetes.io/docs/concepts/overview/working-with-objects/names/#names
type: string
optional:
description: Specify whether the Secret or its key must
be defined
type: boolean
required:
- key
type: object
x-kubernetes-map-type: atomic
required:
- modelId
type: object
modelLocation:
description: Specific location where the model data exists
type: string
modelSourceType:
enum:
- fsx
- s3
- huggingface
- kubernetesVolume
type: string
prefetchEnabled:
default: false
Expand Down Expand Up @@ -5492,6 +5751,44 @@ spec:
- lastUpdated
- name
type: object
dnsStatus:
description: Status of the operator-managed Route53 DNS record
properties:
dnsHealth:
description: 'DNS resolution status: Active, Pending, or Error.'
enum:
- Active
- Pending
- Error
type: string
hostedZoneId:
description: Route53 hosted zone ID.
type: string
lastTransitionTime:
description: When the status last transitioned, used for propagation
timeout.
format: date-time
type: string
managedByOperator:
description: Whether the operator manages this DNS record.
type: boolean
message:
description: Human-readable status or error message.
type: string
previousHostedZoneId:
description: Previous hosted zone ID, retained during domain/zone
changes until cleanup completes.
type: string
previousRecordName:
description: Previous record name, retained during domain/zone
changes until cleanup completes.
type: string
recordName:
description: Route53 record name.
type: string
required:
- managedByOperator
type: object
endpoints:
description: EndpointStatus contains the status of SageMaker endpoints
properties:
Expand Down Expand Up @@ -5801,7 +6098,7 @@ spec:
type: string
serverAddress:
description: Server address for AMP workspace
pattern: ^https:\/\/aps-workspaces\.[a-zA-Z0-9-]+(?:\.[a-zA-Z0-9-]+)*\.amazonaws\.com\/workspaces\/ws-[a-zA-Z0-9-]+\/[a-zA-Z0-9-]+$|^$
pattern: ^https:\/\/aps-workspaces\.[a-zA-Z0-9-]+(?:\.[a-zA-Z0-9-]+)*\.amazonaws\.com\/workspaces\/ws-[a-zA-Z0-9-]+$|^$
type: string
targetValue:
description: Target metric value for scaling
Expand Down
Loading
Loading