Setup the pipelines server

In this section, you will learn how to setup the pipelines server in the Parasol Insurance tenant using ArgoCD.

The data science pipelines server is a component of RHOAI that allows you to create, run, and monitor machine learning pipelines.

Setup the pipelines server (DSPA).

Create a data-science-pipelines directory in the parasol-insurance tenand directory.
Create the base and overlays directories in the data-science-pipelines directory.
Create a directory parasol-insurance-dev in data-science-pipelines/overlays.

In the base directory, create a kustomization.yaml file

Solution

kustomization.yaml

apiVersion: kustomize.config.k8s.io/v1beta1
kind: Kustomization

namespace: parasol-insurance

resources:
  - dspa.yaml

In the base directory, create a dspa.yaml file with the following content:

dspa.yaml

apiVersion: datasciencepipelinesapplications.opendatahub.io/v1alpha1
kind: DataSciencePipelinesApplication
metadata:
  name: dspa
spec:
  apiServer:
    stripEOF: true
    dbConfigConMaxLifetimeSec: 120
    applyTektonCustomResource: true
    deploy: true
    enableSamplePipeline: false
    autoUpdatePipelineDefaultVersion: true
    archiveLogs: false
    terminateStatus: Cancelled
    enableOauth: true
    trackArtifacts: true
    collectMetrics: true
    injectDefaultScript: true
  database:
    disableHealthCheck: false
    mariaDB:
      deploy: true
      pipelineDBName: mlpipeline
      pvcSize: 10Gi
      username: mlpipeline
  dspVersion: v2
  objectStorage:
    disableHealthCheck: false
    enableExternalRoute: false
    externalStorage:
      basePath: ''
      bucket: pipelines
      host: 'minio.object-datastore.svc.cluster.local:9000'
      port: ''
      region: us-east-1
      s3CredentialsSecret:
        accessKey: AWS_ACCESS_KEY_ID
        secretKey: AWS_SECRET_ACCESS_KEY
        secretName: <secret-name>
      scheme: http
  persistenceAgent:
    deploy: true
    numWorkers: 2
  scheduledWorkflow:
    cronScheduleTimezone: UTC
    deploy: true

Your Data Science Project requires a DSPA, so that you can create pipelines.

Before creating the dspa object, your Data Science Project shows a button to Configure pipeline server:

Instead of creating the pipeline server configuration using the RHOAI method that you have seen in previous enablemens, in these steps you are using the GitOps method to create the pipeline server configuration.

In the overlays/parasol-insurance-dev/ directory, create a kustomization.yaml file
Solution
kustomization.yaml
apiVersion: kustomize.config.k8s.io/v1beta1 kind: Kustomization resources: - ../../base
Commit and push the changes to the Git repository.
Wait for ArgoCD to sync the changes.
Validate that the parasol-insurance Data Science project allows to import pipelines under the pipelines tab.