ClearML/clearml-helm-charts
1
0
Fork 0
mirror of https://github.com/clearml/clearml-helm-charts synced 2025-04-17 01:31:13 +00:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

Serving 1.0.0 refactoring (#189)

Browse Source 
* Changed: use dep charts

* Changed: improved ingresses

* Changed: naming management

* Fixed: naming

* Fixed: disable kubestats for prom

* Added: dependencies

* Fixed: typos
This commit is contained in:
Valeriano Manassero 2023-03-20 11:49:42 +01:00 committed by GitHub
parent 957b7b2423
commit cbc1239d10
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
34 changed files with 242 additions and 463 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
.github/workflows/ci.yaml vendored
View File

@ -38,6 +38,10 @@ jobs:
run: helm repo add bitnami https://charts.bitnami.com/bitnami
- name: Add elastic repo
run: helm repo add elastic https://helm.elastic.co
- name: Add prometheus repo
run: helm repo add prometheus https://prometheus-community.github.io/helm-charts
- name: Add grafana repo
run: helm repo add grafana https://grafana.github.io/helm-charts
- name: Run chart-testing (list-changed)
id: list-changed
run: |

12
charts/clearml-serving/Chart.lock Normal file
View File

@ -0,0 +1,12 @@
dependencies:
- name: kafka
repository: https://charts.bitnami.com/bitnami
version: 21.4.0
- name: prometheus
repository: https://prometheus-community.github.io/helm-charts
version: 19.7.2
- name: grafana
repository: https://grafana.github.io/helm-charts
version: 6.52.3
digest: sha256:b28d01875a50b24230ba164d14671225b71d79172192a97e345661e4832f484b
generated: "2023-03-16T09:10:35.77395+01:00"

26
charts/clearml-serving/Chart.yaml
View File

@ -2,9 +2,14 @@ apiVersion: v2
name: clearml-serving
description: ClearML Serving Helm Chart
type: application
version: 0.7.0
version: "1.0.0"
appVersion: "1.2.0"
kubeVersion: ">= 1.19.0-0 < 1.26.0-0"
kubeVersion: ">= 1.21.0-0 < 1.27.0-0"
home: https://clear.ml
icon: https://raw.githubusercontent.com/allegroai/clearml/master/docs/clearml-logo.svg
sources:
- https://github.com/allegroai/clearml-helm-charts
- https://github.com/allegroai/clearml
maintainers:
- name: valeriano-manassero
url: https://github.com/valeriano-manassero
@ -13,3 +18,20 @@ keywords:
- "machine learning"
- mlops
- "model serving"
dependencies:
- name: kafka
version: "21.4.0"
repository: "https://charts.bitnami.com/bitnami"
condition: kafka.enabled
- name: prometheus
version: "19.7.2"
repository: "https://prometheus-community.github.io/helm-charts"
condition: prometheus.enabled
- name: grafana
version: "6.52.3"
repository: "https://grafana.github.io/helm-charts"
condition: grafana.enabled
annotations:
artifacthub.io/changes: |
- kind: changed
description: refactoring

72
charts/clearml-serving/README.md
View File

@ -1,31 +1,71 @@
# clearml-serving
# ClearML Kubernetes Serving
![Version: 0.7.0](https://img.shields.io/badge/Version-0.7.0-informational?style=flat-square) ![Type: application](https://img.shields.io/badge/Type-application-informational?style=flat-square) ![AppVersion: 1.2.0](https://img.shields.io/badge/AppVersion-1.2.0-informational?style=flat-square)
![Version: 1.0.0](https://img.shields.io/badge/Version-1.0.0-informational?style=flat-square) ![Type: application](https://img.shields.io/badge/Type-application-informational?style=flat-square) ![AppVersion: 1.2.0](https://img.shields.io/badge/AppVersion-1.2.0-informational?style=flat-square)
ClearML Serving Helm Chart
**Homepage:** <https://clear.ml>
## Maintainers
| Name | Email | Url |
| ---- | ------ | --- |
| valeriano-manassero | | <https://github.com/valeriano-manassero> |
## Introduction
The **clearml-serving** is the Kubernetes serving for for [ClearML](https://github.com/allegroai/clearml-serving).
It allows you to serve mopdels on a Kubernetes cluster.
# Upgrading Chart
## Upgrades/ Values upgrades
Updating to latest version of this chart can be done in two steps:
```
helm repo update
helm upgrade clearml-serving allegroai/clearml-serving
```
Changing values on existing installation can be done with:
```
helm upgrade clearml-serving allegroai/clearml-serving --version <CURRENT CHART VERSION> -f custom_values.yaml
```
## Source Code
* <https://github.com/allegroai/clearml-helm-charts>
* <https://github.com/allegroai/clearml>
## Requirements
Kubernetes: `>= 1.19.0-0 < 1.26.0-0`
Kubernetes: `>= 1.21.0-0 < 1.27.0-0`
| Repository | Name | Version |
|------------|------|---------|
| https://charts.bitnami.com/bitnami | kafka | 21.4.0 |
| https://grafana.github.io/helm-charts | grafana | 6.52.3 |
| https://prometheus-community.github.io/helm-charts | prometheus | 19.7.2 |
## Values
| Key | Type | Default | Description |
|-----|------|---------|-------------|
| alertmanager | object | `{"affinity":{},"image":{"repository":"prom/alertmanager","tag":"v0.23.0"},"nodeSelector":{},"resources":{},"tolerations":[]}` | Alertmanager generic configigurations |
| clearml | object | `{"apiAccessKey":"ClearML API Access Key","apiHost":"http://clearml-server-apiserver:8008","apiSecretKey":"ClearML API Secret Key","defaultBaseServeUrl":"http://127.0.0.1:8080/serve","filesHost":"http://clearml-server-fileserver:8081","servingTaskId":"ClearML Serving Task ID","webHost":"http://clearml-server-webserver:80"}` | ClearMl generic configurations |
| clearml_serving_inference | object | `{"affinity":{},"autoscaling":{"enabled":false,"maxReplicas":11,"minReplicas":1,"targetCPU":50,"targetMemory":50},"extraPythonPackages":[],"image":{"repository":"allegroai/clearml-serving-inference","tag":"1.2.0"},"ingress":{"annotations":{},"enabled":false,"hostName":"serving.clearml.127-0-0-1.nip.io","path":"/","tlsSecretName":""},"nodeSelector":{},"resources":{},"tolerations":[]}` | ClearML serving inference configurations |
| clearml_serving_inference | object | `{"affinity":{},"autoscaling":{"enabled":false,"maxReplicas":11,"minReplicas":1,"targetCPU":50,"targetMemory":50},"extraPythonPackages":[],"image":{"repository":"allegroai/clearml-serving-inference","tag":"1.2.0"},"ingress":{"annotations":{},"enabled":false,"hostName":"serving.clearml.127-0-0-1.nip.io","ingressClassName":"","path":"/","tlsSecretName":""},"nodeSelector":{},"resources":{},"tolerations":[]}` | ClearML serving inference configurations |
| clearml_serving_inference.affinity | object | `{}` | Affinity configuration |
| clearml_serving_inference.autoscaling | object | `{"enabled":false,"maxReplicas":11,"minReplicas":1,"targetCPU":50,"targetMemory":50}` | Autoscaling configuration |
| clearml_serving_inference.extraPythonPackages | list | `[]` | Extra Python Packages to be installed in running pods |
| clearml_serving_inference.image | object | `{"repository":"allegroai/clearml-serving-inference","tag":"1.2.0"}` | Container Image |
| clearml_serving_inference.ingress | object | `{"annotations":{},"enabled":false,"hostName":"serving.clearml.127-0-0-1.nip.io","path":"/","tlsSecretName":""}` | Ingress exposing configurations |
| clearml_serving_inference.ingress | object | `{"annotations":{},"enabled":false,"hostName":"serving.clearml.127-0-0-1.nip.io","ingressClassName":"","path":"/","tlsSecretName":""}` | Ingress exposing configurations |
| clearml_serving_inference.ingress.annotations | object | `{}` | Ingress annotations |
| clearml_serving_inference.ingress.enabled | bool | `false` | Enable/Disable ingress |
| clearml_serving_inference.ingress.hostName | string | `"serving.clearml.127-0-0-1.nip.io"` | Ingress hostname domain |
| clearml_serving_inference.ingress.ingressClassName | string | `""` | ClassName (must be defined if no default ingressClassName is available) |
| clearml_serving_inference.ingress.path | string | `"/"` | Ingress root path url |
| clearml_serving_inference.ingress.tlsSecretName | string | `""` | Reference to secret containing TLS certificate. If set, it enables HTTPS on ingress rule. |
| clearml_serving_inference.nodeSelector | object | `{}` | Node Selector configuration |
| clearml_serving_inference.resources | object | `{}` | Pod resources definition |
| clearml_serving_inference.tolerations | list | `[]` | Tolerations configuration |
@ -36,20 +76,22 @@ Kubernetes: `>= 1.19.0-0 < 1.26.0-0`
| clearml_serving_statistics.nodeSelector | object | `{}` | Node Selector configuration |
| clearml_serving_statistics.resources | object | `{}` | Pod resources definition |
| clearml_serving_statistics.tolerations | list | `[]` | Tolerations configuration |
| clearml_serving_triton | object | `{"affinity":{},"autoscaling":{"enabled":false,"maxReplicas":11,"minReplicas":1,"targetCPU":50,"targetMemory":50},"enabled":true,"extraPythonPackages":[],"image":{"repository":"allegroai/clearml-serving-triton","tag":"1.2.0-22.07"},"ingress":{"annotations":{},"enabled":false,"hostName":"serving-grpc.clearml.127-0-0-1.nip.io","path":"/","tlsSecretName":""},"nodeSelector":{},"resources":{},"tolerations":[]}` | ClearML serving Triton configurations |
| clearml_serving_triton | object | `{"affinity":{},"autoscaling":{"enabled":false,"maxReplicas":11,"minReplicas":1,"targetCPU":50,"targetMemory":50},"enabled":true,"extraPythonPackages":[],"image":{"repository":"allegroai/clearml-serving-triton","tag":"1.2.0-22.07"},"ingress":{"annotations":{},"enabled":false,"hostName":"serving-grpc.clearml.127-0-0-1.nip.io","ingressClassName":"","path":"/","tlsSecretName":""},"nodeSelector":{},"resources":{},"tolerations":[]}` | ClearML serving Triton configurations |
| clearml_serving_triton.affinity | object | `{}` | Affinity configuration |
| clearml_serving_triton.autoscaling | object | `{"enabled":false,"maxReplicas":11,"minReplicas":1,"targetCPU":50,"targetMemory":50}` | Autoscaling configuration |
| clearml_serving_triton.enabled | bool | `true` | Triton pod creation enable/disable |
| clearml_serving_triton.extraPythonPackages | list | `[]` | Extra Python Packages to be installed in running pods |
| clearml_serving_triton.image | object | `{"repository":"allegroai/clearml-serving-triton","tag":"1.2.0-22.07"}` | Container Image |
| clearml_serving_triton.ingress | object | `{"annotations":{},"enabled":false,"hostName":"serving-grpc.clearml.127-0-0-1.nip.io","path":"/","tlsSecretName":""}` | Ingress exposing configurations |
| clearml_serving_triton.ingress | object | `{"annotations":{},"enabled":false,"hostName":"serving-grpc.clearml.127-0-0-1.nip.io","ingressClassName":"","path":"/","tlsSecretName":""}` | Ingress exposing configurations |
| clearml_serving_triton.ingress.annotations | object | `{}` | Ingress annotations |
| clearml_serving_triton.ingress.enabled | bool | `false` | Enable/Disable ingress |
| clearml_serving_triton.ingress.hostName | string | `"serving-grpc.clearml.127-0-0-1.nip.io"` | Ingress hostname domain |
| clearml_serving_triton.ingress.ingressClassName | string | `""` | ClassName (must be defined if no default ingressClassName is available) |
| clearml_serving_triton.ingress.path | string | `"/"` | Ingress root path url |
| clearml_serving_triton.ingress.tlsSecretName | string | `""` | Reference to secret containing TLS certificate. If set, it enables HTTPS on ingress rule. |
| clearml_serving_triton.nodeSelector | object | `{}` | Node Selector configuration |
| clearml_serving_triton.resources | object | `{}` | Pod resources definition |
| clearml_serving_triton.tolerations | list | `[]` | Tolerations configuration |
| grafana | object | `{"affinity":{},"image":{"repository":"grafana/grafana","tag":"8.4.4-ubuntu"},"ingress":{"annotations":{},"enabled":false,"hostName":"serving-grafana.clearml.127-0-0-1.nip.io","path":"/","tlsSecretName":""},"nodeSelector":{},"resources":{},"tolerations":[]}` | Grafana generic configigurations |
| kafka | object | `{"affinity":{},"image":{"repository":"bitnami/kafka","tag":"3.1.0"},"nodeSelector":{},"resources":{},"tolerations":[]}` | Kafka generic configigurations |
| prometheus | object | `{"affinity":{},"image":{"repository":"prom/prometheus","tag":"v2.34.0"},"nodeSelector":{},"resources":{},"tolerations":[]}` | Prometheus generic configigurations |
| zookeeper | object | `{"affinity":{},"image":{"repository":"bitnami/zookeeper","tag":"3.7.0"},"nodeSelector":{},"resources":{},"tolerations":[]}` | Zookeeper generic configigurations |
----------------------------------------------
Autogenerated from chart metadata using [helm-docs v1.11.0](https://github.com/norwoodj/helm-docs/releases/v1.11.0)
| grafana | object | `{"adminPassword":"clearml","adminUser":"admin","datasources":{"datasources.yaml":{"apiVersion":1,"datasources":[{"access":"proxy","isDefault":true,"name":"Prometheus","type":"prometheus","url":"http://{{ .Release.Name }}-prometheus-server"}]}},"enabled":true}` | Configuration from https://github.com/grafana/helm-charts/blob/main/charts/grafana/values.yaml |
| kafka | object | `{"enabled":true}` | Configuration from https://github.com/bitnami/charts/blob/main/bitnami/kafka/values.yaml |
| prometheus | object | `{"enabled":true,"extraScrapeConfigs":"- job_name: \"{{ .Release.Name }}-stats\"\n static_configs:\n - targets:\n - \"{{ .Release.Name }}-statistics:9999\"\n","kube-state-metrics":{"enabled":false},"prometheus-node-exporter":{"enabled":false},"prometheus-pushgateway":{"enabled":false},"serverFiles":{"prometheus.yml":{"scrape_configs":[{"job_name":"prometheus","static_configs":[{"targets":["localhost:9090"]}]}]}}}` | Configuration from https://github.com/prometheus-community/helm-charts/blob/main/charts/prometheus/values.yaml |

38
charts/clearml-serving/README.md.gotmpl Normal file
View File

@ -0,0 +1,38 @@
# ClearML Kubernetes Serving
{{ template "chart.deprecationWarning" . }}
{{ template "chart.badgesSection" . }}
{{ template "chart.description" . }}
{{ template "chart.homepageLine" . }}
{{ template "chart.maintainersSection" . }}
## Introduction
The **clearml-serving** is the Kubernetes serving for for [ClearML](https://github.com/allegroai/clearml-serving).
It allows you to serve mopdels on a Kubernetes cluster.
# Upgrading Chart
## Upgrades/ Values upgrades
Updating to latest version of this chart can be done in two steps:
```
helm repo update
helm upgrade clearml-serving allegroai/clearml-serving
```
Changing values on existing installation can be done with:
```
helm upgrade clearml-serving allegroai/clearml-serving --version <CURRENT CHART VERSION> -f custom_values.yaml
```
{{ template "chart.sourcesSection" . }}
{{ template "chart.requirementsSection" . }}
{{ template "chart.valuesSection" . }}

BIN
charts/clearml-serving/charts/grafana-6.52.3.tgz Normal file
View File

Binary file not shown.

BIN
charts/clearml-serving/charts/kafka-21.4.0.tgz Normal file
View File

Binary file not shown.

BIN
charts/clearml-serving/charts/prometheus-19.7.2.tgz Normal file
View File

Binary file not shown.

20
charts/clearml-serving/templates/_helpers.tpl
View File

@ -1,7 +1,7 @@
{{/*
Expand the name of the chart.
*/}}
{{- define "clearml-serving.name" -}}
{{- define "clearmlServing.name" -}}
{{- default .Chart.Name .Values.nameOverride | trunc 63 | trimSuffix "-" }}
{{- end }}
@ -10,7 +10,7 @@ Create a default fully qualified app name.
We truncate at 63 chars because some Kubernetes name fields are limited to this (by the DNS naming spec).
If release name contains chart name it will be used as a full name.
*/}}
{{- define "clearml-serving.fullname" -}}
{{- define "clearmlServing.fullname" -}}
{{- if .Values.fullnameOverride }}
{{- .Values.fullnameOverride | trunc 63 | trimSuffix "-" }}
{{- else }}
@ -26,16 +26,16 @@ If release name contains chart name it will be used as a full name.
{{/*
Create chart name and version as used by the chart label.
*/}}
{{- define "clearml-serving.chart" -}}
{{- define "clearmlServing.chart" -}}
{{- printf "%s-%s" .Chart.Name .Chart.Version | replace "+" "_" | trunc 63 | trimSuffix "-" }}
{{- end }}
{{/*
Common labels
*/}}
{{- define "clearml-serving.labels" -}}
helm.sh/chart: {{ include "clearml-serving.chart" . }}
{{ include "clearml-serving.selectorLabels" . }}
{{- define "clearmlServing.labels" -}}
helm.sh/chart: {{ include "clearmlServing.chart" . }}
{{ include "clearmlServing.selectorLabels" . }}
{{- if .Chart.AppVersion }}
app.kubernetes.io/version: {{ .Chart.AppVersion | quote }}
{{- end }}
@ -45,17 +45,17 @@ app.kubernetes.io/managed-by: {{ .Release.Service }}
{{/*
Selector labels
*/}}
{{- define "clearml-serving.selectorLabels" -}}
app.kubernetes.io/name: {{ include "clearml-serving.name" . }}
{{- define "clearmlServing.selectorLabels" -}}
app.kubernetes.io/name: {{ include "clearmlServing.name" . }}
app.kubernetes.io/instance: {{ .Release.Name }}
{{- end }}
{{/*
Create the name of the service account to use
*/}}
{{- define "clearml-serving.serviceAccountName" -}}
{{- define "clearmlServing.serviceAccountName" -}}
{{- if .Values.serviceAccount.create }}
{{- default (include "clearml-serving.fullname" .) .Values.serviceAccount.name }}
{{- default (include "clearmlServing.fullname" .) .Values.serviceAccount.name }}
{{- else }}
{{- default "default" .Values.serviceAccount.name }}
{{- end }}

27
charts/clearml-serving/templates/alertmanager-deployment.yaml
View File

@ -1,27 +0,0 @@
apiVersion: apps/v1
kind: Deployment
metadata:
annotations: {}
labels:
clearml.serving.service: alertmanager
name: alertmanager
spec:
replicas: 1
selector:
matchLabels:
clearml.serving.service: alertmanager
strategy: {}
template:
metadata:
annotations: {}
labels:
clearml.serving.network/clearml-serving-backend: "true"
clearml.serving.service: alertmanager
spec:
containers:
- image: "{{ .Values.alertmanager.image.repository }}:{{ .Values.alertmanager.image.tag }}"
name: clearml-serving-alertmanager
ports:
- containerPort: 9093
resources: {}
restartPolicy: Always

14
charts/clearml-serving/templates/alertmanager-service.yaml
View File

@ -1,14 +0,0 @@
apiVersion: v1
kind: Service
metadata:
annotations: {}
labels:
clearml.serving.service: alertmanager
name: clearml-serving-alertmanager
spec:
ports:
- name: "9093"
port: 9093
targetPort: 9093
selector:
clearml.serving.service: alertmanager

2
charts/clearml-serving/templates/clearml-serving-backend-networkpolicy.yaml
View File

@ -1,7 +1,7 @@
apiVersion: networking.k8s.io/v1
kind: NetworkPolicy
metadata:
name: clearml-serving-backend
name: {{ include "clearmlServing.fullname" . }}-backend
spec:
ingress:
- from:

14
charts/clearml-serving/templates/clearml-serving-inference-deployment.yaml
View File

@ -3,20 +3,20 @@ kind: Deployment
metadata:
annotations: {}
labels:
clearml.serving.service: clearml-serving-inference
name: clearml-serving-inference
clearml.serving.service: {{ include "clearmlServing.fullname" . }}-inference
name: {{ include "clearmlServing.fullname" . }}-inference
spec:
replicas: 1
selector:
matchLabels:
clearml.serving.service: clearml-serving-inference
clearml.serving.service: {{ include "clearmlServing.fullname" . }}-inference
strategy: {}
template:
metadata:
annotations: {}
labels:
clearml.serving.network/clearml-serving-backend: "true"
clearml.serving.service: clearml-serving-inference
clearml.serving.service: {{ include "clearmlServing.fullname" . }}-inference
spec:
containers:
- env:
@ -31,14 +31,14 @@ spec:
- name: CLEARML_WEB_HOST
value: "{{ .Values.clearml.webHost }}"
- name: CLEARML_DEFAULT_KAFKA_SERVE_URL
value: clearml-serving-kafka:9092
value: {{ include "clearmlServing.fullname" . }}-kafka:9092
- name: CLEARML_SERVING_POLL_FREQ
value: "1.0"
- name: CLEARML_DEFAULT_BASE_SERVE_URL
value: "{{ .Values.clearml.defaultBaseServeUrl }}"
- name: CLEARML_DEFAULT_TRITON_GRPC_ADDR
{{- if .Values.clearml_serving_triton.enabled }}
value: "clearml-serving-triton:8001"
value: "{{ include "clearmlServing.fullname" . }}-triton:8001"
{{- else }}
value: ""
{{- end }}
@ -55,7 +55,7 @@ spec:
value: '{{ join " " .Values.clearml_serving_inference.extraPythonPackages }}'
{{- end }}
image: "{{ .Values.clearml_serving_inference.image.repository }}:{{ .Values.clearml_serving_inference.image.tag }}"
name: clearml-serving-inference
name: {{ include "clearmlServing.fullname" . }}-inference
ports:
- containerPort: 8080
resources: {}

6
charts/clearml-serving/templates/clearml-serving-inference-hpa.yaml
View File

@ -2,16 +2,16 @@
apiVersion: {{ include "common.capabilities.hpa.apiVersion" ( dict "context" $ ) }}
kind: HorizontalPodAutoscaler
metadata:
name: clearml-serving-inference-hpa
name: {{ include "clearmlServing.fullname" . }}-inference-hpa
namespace: {{ .Release.Namespace | quote }}
annotations: {}
labels:
clearml.serving.service: clearml-serving-inference
clearml.serving.service: {{ include "clearmlServing.fullname" . }}-inference
spec:
scaleTargetRef:
apiVersion: "apps/v1"
kind: Deployment
name: clearml-serving-inference
name: {{ include "clearmlServing.fullname" . }}-inference
minReplicas: {{ .Values.clearml_serving_inference.autoscaling.minReplicas }}
maxReplicas: {{ .Values.clearml_serving_inference.autoscaling.maxReplicas }}
metrics:

8
charts/clearml-serving/templates/clearml-serving-inference-ingress.yaml
View File

@ -8,9 +8,9 @@ apiVersion: extensions/v1beta1
{{- end }}
kind: Ingress
metadata:
name: clearml-serving-inference
name: {{ include "clearmlServing.fullname" . }}-inference
labels:
clearml.serving.service: clearml-serving-inference
clearml.serving.service: {{ include "clearmlServing.fullname" . }}-inference
annotations:
{{- toYaml .Values.clearml_serving_inference.ingress.annotations | nindent 4 }}
spec:
@ -29,12 +29,12 @@ spec:
pathType: Prefix
backend:
service:
name: clearml-serving-inference
name: {{ include "clearmlServing.fullname" . }}-inference
port:
number: 8080
{{ else }}
backend:
serviceName: clearml-serving-inference
servicename: {{ include "clearmlServing.fullname" . }}-inference
servicePort: 8080
{{ end }}
{{- end }}

6
charts/clearml-serving/templates/clearml-serving-inference-service.yaml
View File

@ -3,12 +3,12 @@ kind: Service
metadata:
annotations: {}
labels:
clearml.serving.service: clearml-serving-inference
name: clearml-serving-inference
clearml.serving.service: {{ include "clearmlServing.fullname" . }}-inference
name: {{ include "clearmlServing.fullname" . }}-inference
spec:
ports:
- name: "8080"
port: 8080
targetPort: 8080
selector:
clearml.serving.service: clearml-serving-inference
clearml.serving.service: {{ include "clearmlServing.fullname" . }}-inference

12
charts/clearml-serving/templates/clearml-serving-statistics-deployment.yaml
View File

@ -3,20 +3,20 @@ kind: Deployment
metadata:
annotations: {}
labels:
clearml.serving.service: clearml-serving-statistics
name: clearml-serving-statistics
clearml.serving.service: {{ include "clearmlServing.fullname" . }}-statistics
name: {{ include "clearmlServing.fullname" . }}-statistics
spec:
replicas: 1
selector:
matchLabels:
clearml.serving.service: clearml-serving-statistics
clearml.serving.service: {{ include "clearmlServing.fullname" . }}-statistics
strategy: {}
template:
metadata:
annotations: {}
labels:
clearml.serving.network/clearml-serving-backend: "true"
clearml.serving.service: clearml-serving-statistics
clearml.serving.service: {{ include "clearmlServing.fullname" . }}-statistics
spec:
containers:
- env:
@ -31,7 +31,7 @@ spec:
- name: CLEARML_WEB_HOST
value: "{{ .Values.clearml.webHost }}"
- name: CLEARML_DEFAULT_KAFKA_SERVE_URL
value: clearml-serving-kafka:9092
value: {{ include "clearmlServing.fullname" . }}-kafka:9092
- name: CLEARML_SERVING_POLL_FREQ
value: "1.0"
- name: CLEARML_SERVING_TASK_ID
@ -41,7 +41,7 @@ spec:
value: '{{ join " " .Values.clearml_serving_statistics.extraPythonPackages }}'
{{- end }}
image: "{{ .Values.clearml_serving_statistics.image.repository }}:{{ .Values.clearml_serving_statistics.image.tag }}"
name: clearml-serving-statistics
name: {{ include "clearmlServing.fullname" . }}-statistics
ports:
- containerPort: 9999
resources: {}

6
charts/clearml-serving/templates/clearml-serving-statistics-service.yaml
View File

@ -3,12 +3,12 @@ kind: Service
metadata:
annotations: {}
labels:
clearml.serving.service: clearml-serving-statistics
name: clearml-serving-statistics
clearml.serving.service: {{ include "clearmlServing.fullname" . }}-statistics
name: {{ include "clearmlServing.fullname" . }}-statistics
spec:
ports:
- name: "9999"
port: 9999
targetPort: 9999
selector:
clearml.serving.service: clearml-serving-statistics
clearml.serving.service: {{ include "clearmlServing.fullname" . }}-statistics

10
charts/clearml-serving/templates/clearml-serving-triton-deployment.yaml
View File

@ -4,20 +4,20 @@ kind: Deployment
metadata:
annotations: {}
labels:
clearml.serving.service: clearml-serving-triton
name: clearml-serving-triton
clearml.serving.service: {{ include "clearmlServing.fullname" . }}-triton
name: {{ include "clearmlServing.fullname" . }}-triton
spec:
replicas: 1
selector:
matchLabels:
clearml.serving.service: clearml-serving-triton
clearml.serving.service: {{ include "clearmlServing.fullname" . }}-triton
strategy: {}
template:
metadata:
annotations: {}
labels:
clearml.serving.network/clearml-serving-backend: "true"
clearml.serving.service: clearml-serving-triton
clearml.serving.service: {{ include "clearmlServing.fullname" . }}-triton
spec:
containers:
- env:
@ -42,7 +42,7 @@ spec:
value: '{{ join " " .Values.clearml_serving_triton.extraPythonPackages }}'
{{- end }}
image: "{{ .Values.clearml_serving_triton.image.repository }}:{{ .Values.clearml_serving_triton.image.tag }}"
name: clearml-serving-triton
name: {{ include "clearmlServing.fullname" . }}-triton
ports:
- containerPort: 8001
resources: {}

6
charts/clearml-serving/templates/clearml-serving-triton-hpa.yaml
View File

@ -2,16 +2,16 @@
apiVersion: {{ include "common.capabilities.hpa.apiVersion" ( dict "context" $ ) }}
kind: HorizontalPodAutoscaler
metadata:
name: clearml-serving-triton-hpa
name: {{ include "clearmlServing.fullname" . }}-triton-hpa
namespace: {{ .Release.Namespace | quote }}
annotations: {}
labels:
clearml.serving.service: clearml-serving-triton
clearml.serving.service: {{ include "clearmlServing.fullname" . }}-triton
spec:
scaleTargetRef:
apiVersion: "apps/v1"
kind: Deployment
name: clearml-serving-triton
name: {{ include "clearmlServing.fullname" . }}-triton
minReplicas: {{ .Values.clearml_serving_triton.autoscaling.minReplicas }}
maxReplicas: {{ .Values.clearml_serving_triton.autoscaling.maxReplicas }}
metrics:

8
charts/clearml-serving/templates/clearml-serving-triton-ingress.yaml
View File

@ -9,9 +9,9 @@ apiVersion: extensions/v1beta1
{{- end }}
kind: Ingress
metadata:
name: clearml-serving-triton
name: {{ include "clearmlServing.fullname" . }}-triton
labels:
clearml.serving.service: clearml-serving-triton
clearml.serving.service: {{ include "clearmlServing.fullname" . }}-triton
annotations:
{{- toYaml .Values.clearml_serving_triton.ingress.annotations | nindent 4 }}
spec:
@ -30,12 +30,12 @@ spec:
pathType: Prefix
backend:
service:
name: clearml-serving-triton
name: {{ include "clearmlServing.fullname" . }}-triton
port:
number: 8001
{{ else }}
backend:
serviceName: clearml-serving-triton
servicename: {{ include "clearmlServing.fullname" . }}-triton
servicePort: 8001
{{ end }}
{{- end }}

6
charts/clearml-serving/templates/clearml-serving-triton-service.yaml
View File

@ -4,13 +4,13 @@ kind: Service
metadata:
annotations: {}
labels:
clearml.serving.service: clearml-serving-triton
name: clearml-serving-triton
clearml.serving.service: {{ include "clearmlServing.fullname" . }}-triton
name: {{ include "clearmlServing.fullname" . }}-triton
spec:
ports:
- name: "8001"
port: 8001
targetPort: 8001
selector:
clearml.serving.service: clearml-serving-triton
clearml.serving.service: {{ include "clearmlServing.fullname" . }}-triton
{{ end }}

14
charts/clearml-serving/templates/grafana-config-secret.yaml
View File

@ -1,14 +0,0 @@
apiVersion: v1
kind: Secret
metadata:
name: grafana-config
stringData:
datasource.yaml: |-
apiVersion: 1
datasources:
- name: Prometheus
type: prometheus
# Access mode - proxy (server in the UI) or direct (browser in the UI).
access: proxy
url: http://clearml-serving-prometheus:9090

35
charts/clearml-serving/templates/grafana-deployment.yaml
View File

@ -1,35 +0,0 @@
apiVersion: apps/v1
kind: Deployment
metadata:
annotations: {}
labels:
clearml.serving.service: grafana
name: grafana
spec:
replicas: 1
selector:
matchLabels:
clearml.serving.service: grafana
strategy:
type: Recreate
template:
metadata:
annotations: {}
labels:
clearml.serving.network/clearml-serving-backend: "true"
clearml.serving.service: grafana
spec:
containers:
- image: "{{ .Values.grafana.image.repository }}:{{ .Values.grafana.image.tag }}"
name: clearml-serving-grafana
ports:
- containerPort: 3000
resources: {}
volumeMounts:
- mountPath: /etc/grafana/provisioning/datasources/
name: grafana-conf
restartPolicy: Always
volumes:
- name: grafana-conf
secret:
secretName: grafana-config

40
charts/clearml-serving/templates/grafana-ingress.yaml
View File

@ -1,40 +0,0 @@
{{- if .Values.grafana.ingress.enabled -}}
{{- if semverCompare ">=1.19-0" .Capabilities.KubeVersion.GitVersion -}}
apiVersion: networking.k8s.io/v1
{{- else if semverCompare ">=1.14-0" .Capabilities.KubeVersion.GitVersion -}}
apiVersion: networking.k8s.io/v1beta1
{{- else -}}
apiVersion: extensions/v1beta1
{{- end }}
kind: Ingress
metadata:
name: clearml-serving-grafana
labels:
clearml.serving.service: clearml-serving-grafana
annotations:
{{- toYaml .Values.grafana.ingress.annotations | nindent 4 }}
spec:
{{- if .Values.grafana.ingress.tlsSecretName }}
tls:
- hosts:
- {{ .Values.grafana.ingress.hostName }}
secretName: {{ .Values.grafana.ingress.tlsSecretName }}
{{- end }}
rules:
- host: {{ .Values.grafana.ingress.hostName }}
http:
paths:
- path: {{ .Values.grafana.ingress.path }}
{{ if semverCompare ">=1.19-0" .Capabilities.KubeVersion.GitVersion }}
pathType: Prefix
backend:
service:
name: clearml-serving-grafana
port:
number: 3000
{{ else }}
backend:
serviceName: clearml-serving-grafana
servicePort: 3000
{{ end }}
{{- end }}

14
charts/clearml-serving/templates/grafana-service.yaml
View File

@ -1,14 +0,0 @@
apiVersion: v1
kind: Service
metadata:
annotations: {}
labels:
clearml.serving.service: grafana
name: clearml-serving-grafana
spec:
ports:
- name: "3000"
port: 3000
targetPort: 3000
selector:
clearml.serving.service: grafana

40
charts/clearml-serving/templates/kafka-deployment.yaml
View File

@ -1,40 +0,0 @@
apiVersion: apps/v1
kind: Deployment
metadata:
annotations: {}
labels:
clearml.serving.service: kafka
name: kafka
spec:
replicas: 1
selector:
matchLabels:
clearml.serving.service: kafka
strategy: {}
template:
metadata:
annotations: {}
labels:
clearml.serving.network/clearml-serving-backend: "true"
clearml.serving.service: kafka
spec:
containers:
- env:
- name: ALLOW_PLAINTEXT_LISTENER
value: "yes"
- name: KAFKA_BROKER_ID
value: "1"
- name: KAFKA_CFG_ADVERTISED_LISTENERS
value: PLAINTEXT://clearml-serving-kafka:9092
- name: KAFKA_CFG_LISTENERS
value: PLAINTEXT://0.0.0.0:9092
- name: KAFKA_CFG_ZOOKEEPER_CONNECT
value: clearml-serving-zookeeper:2181
- name: KAFKA_CREATE_TOPICS
value: '"topic_test:1:1"'
image: "{{ .Values.kafka.image.repository }}:{{ .Values.kafka.image.tag }}"
name: clearml-serving-kafka
ports:
- containerPort: 9092
resources: {}
restartPolicy: Always

14
charts/clearml-serving/templates/kafka-service.yaml
View File

@ -1,14 +0,0 @@
apiVersion: v1
kind: Service
metadata:
annotations: {}
labels:
clearml.serving.service: kafka
name: clearml-serving-kafka
spec:
ports:
- name: "9092"
port: 9092
targetPort: 9092
selector:
clearml.serving.service: kafka

28
charts/clearml-serving/templates/prometheus-config-secret.yaml
View File

@ -1,28 +0,0 @@
apiVersion: v1
kind: Secret
metadata:
name: prometheus-config
stringData:
prometheus.yml: |-
global:
scrape_interval: "15s" # By default, scrape targets every 15 seconds.
evaluation_interval: 15s # By default, scrape targets every 15 seconds.
external_labels:
monitor: 'clearml-serving'
scrape_configs:
# The job name is added as a label `job=<job_name>` to any timeseries scraped from this config.
- job_name: 'prometheus'
scrape_interval: 5s
static_configs:
- targets: ['localhost:9090']
# The job name is added as a label `job=<job_name>` to any timeseries scraped from this config.
- job_name: 'clearml-inference-stats'
scrape_interval: 5s
static_configs:
- targets: ['clearml-serving-statistics:9999']

42
charts/clearml-serving/templates/prometheus-deployment.yaml
View File

@ -1,42 +0,0 @@
apiVersion: apps/v1
kind: Deployment
metadata:
annotations: {}
labels:
clearml.serving.service: prometheus
name: prometheus
spec:
replicas: 1
selector:
matchLabels:
clearml.serving.service: prometheus
strategy:
type: Recreate
template:
metadata:
annotations: {}
labels:
clearml.serving.network/clearml-serving-backend: "true"
clearml.serving.service: prometheus
spec:
containers:
- args:
- --config.file=/mnt/prometheus.yml
- --storage.tsdb.path=/prometheus
- --web.console.libraries=/etc/prometheus/console_libraries
- --web.console.templates=/etc/prometheus/consoles
- --storage.tsdb.retention.time=200h
- --web.enable-lifecycle
image: "{{ .Values.prometheus.image.repository }}:{{ .Values.prometheus.image.tag }}"
name: clearml-serving-prometheus
ports:
- containerPort: 9090
resources: {}
volumeMounts:
- mountPath: /mnt
name: prometheus-conf
restartPolicy: Always
volumes:
- name: prometheus-conf
secret:
secretName: prometheus-config

14
charts/clearml-serving/templates/prometheus-service.yaml
View File

@ -1,14 +0,0 @@
apiVersion: v1
kind: Service
metadata:
annotations: {}
labels:
clearml.serving.service: prometheus
name: clearml-serving-prometheus
spec:
ports:
- name: "9090"
port: 9090
targetPort: 9090
selector:
clearml.serving.service: prometheus

30
charts/clearml-serving/templates/zookeeper-deployment.yaml
View File

@ -1,30 +0,0 @@
apiVersion: apps/v1
kind: Deployment
metadata:
annotations: {}
labels:
clearml.serving.service: zookeeper
name: zookeeper
spec:
replicas: 1
selector:
matchLabels:
clearml.serving.service: zookeeper
strategy: {}
template:
metadata:
annotations: {}
labels:
clearml.serving.network/clearml-serving-backend: "true"
clearml.serving.service: zookeeper
spec:
containers:
- env:
- name: ALLOW_ANONYMOUS_LOGIN
value: "yes"
image: "{{ .Values.zookeeper.image.repository }}:{{ .Values.zookeeper.image.tag }}"
name: clearml-serving-zookeeper
ports:
- containerPort: 2181
resources: {}
restartPolicy: Always

14
charts/clearml-serving/templates/zookeeper-service.yaml
View File

@ -1,14 +0,0 @@
apiVersion: v1
kind: Service
metadata:
annotations: {}
labels:
clearml.serving.service: zookeeper
name: clearml-serving-zookeeper
spec:
ports:
- name: "2181"
port: 2181
targetPort: 2181
selector:
clearml.serving.service: zookeeper

123
charts/clearml-serving/values.yaml
View File

@ -8,62 +8,6 @@ clearml:
defaultBaseServeUrl: http://127.0.0.1:8080/serve
servingTaskId: "ClearML Serving Task ID"
# -- Zookeeper generic configigurations
zookeeper:
image:
repository: "bitnami/zookeeper"
tag: "3.7.0"
nodeSelector: {}
tolerations: []
affinity: {}
resources: {}
# -- Kafka generic configigurations
kafka:
image:
repository: "bitnami/kafka"
tag: "3.1.0"
nodeSelector: {}
tolerations: []
affinity: {}
resources: {}
# -- Prometheus generic configigurations
prometheus:
image:
repository: "prom/prometheus"
tag: "v2.34.0"
nodeSelector: {}
tolerations: []
affinity: {}
resources: {}
# -- Grafana generic configigurations
grafana:
image:
repository: "grafana/grafana"
tag: "8.4.4-ubuntu"
nodeSelector: {}
tolerations: []
affinity: {}
resources: {}
ingress:
enabled: false
hostName: "serving-grafana.clearml.127-0-0-1.nip.io"
tlsSecretName: ""
annotations: {}
path: "/"
# -- Alertmanager generic configigurations
alertmanager:
image:
repository: "prom/alertmanager"
tag: "v0.23.0"
nodeSelector: {}
tolerations: []
affinity: {}
resources: {}
# -- ClearML serving statistics configurations
clearml_serving_statistics:
# -- Container Image
@ -110,10 +54,17 @@ clearml_serving_inference:
targetMemory: 50
# -- Ingress exposing configurations
ingress:
# -- Enable/Disable ingress
enabled: false
# -- ClassName (must be defined if no default ingressClassName is available)
ingressClassName: ""
# -- Ingress hostname domain
hostName: "serving.clearml.127-0-0-1.nip.io"
# -- Reference to secret containing TLS certificate. If set, it enables HTTPS on ingress rule.
tlsSecretName: ""
# -- Ingress annotations
annotations: {}
# -- Ingress root path url
path: "/"
# -- ClearML serving Triton configurations
@ -145,20 +96,56 @@ clearml_serving_triton:
targetMemory: 50
# -- Ingress exposing configurations
ingress:
# -- Enable/Disable ingress
enabled: false
# -- ClassName (must be defined if no default ingressClassName is available)
ingressClassName: ""
# -- Ingress hostname domain
hostName: "serving-grpc.clearml.127-0-0-1.nip.io"
# -- Reference to secret containing TLS certificate. If set, it enables HTTPS on ingress rule.
tlsSecretName: ""
# -- Ingress annotations
annotations: {}
# # Example for AWS ALB
# kubernetes.io/ingress.class: alb
# alb.ingress.kubernetes.io/backend-protocol: HTTP
# alb.ingress.kubernetes.io/backend-protocol-version: GRPC
# alb.ingress.kubernetes.io/certificate-arn: <cerntificate arn>
# alb.ingress.kubernetes.io/ssl-redirect: '443'
# alb.ingress.kubernetes.io/listen-ports: '[{"HTTPS":443}]'
# alb.ingress.kubernetes.io/target-type: ip
#
# # Example for NNGINX ingress controller
# nginx.ingress.kubernetes.io/ssl-redirect: "true"
# nginx.ingress.kubernetes.io/backend-protocol: "GRPC"
# -- Ingress root path url
path: "/"
# -- Configuration from https://github.com/bitnami/charts/blob/main/bitnami/kafka/values.yaml
kafka:
enabled: true
# -- Configuration from https://github.com/prometheus-community/helm-charts/blob/main/charts/prometheus/values.yaml
prometheus:
enabled: true
kube-state-metrics:
enabled: false
prometheus-node-exporter:
enabled: false
prometheus-pushgateway:
enabled: false
serverFiles:
prometheus.yml:
scrape_configs:
- job_name: prometheus
static_configs:
- targets:
- localhost:9090
extraScrapeConfigs: |
- job_name: "{{ .Release.Name }}-stats"
static_configs:
- targets:
- "{{ .Release.Name }}-statistics:9999"
# -- Configuration from https://github.com/grafana/helm-charts/blob/main/charts/grafana/values.yaml
grafana:
enabled: true
adminUser: admin
adminPassword: clearml
datasources:
datasources.yaml:
apiVersion: 1
datasources:
- name: Prometheus
type: prometheus
url: "http://{{ .Release.Name }}-prometheus-server"
access: proxy
isDefault: true