Co-authored-by: Matthias Kretschmann <m@kretschmann.io>
5.8 KiB
title | description |
---|---|
Set Up a Compute-to-Data Environment | Set Up a Compute-to-Data enviroment. |
Requirements
At the time of writing, we need the following: First, create a folder with the following structure:
- working Barge. For this setup, we will asume the Barge is installed in /ocean/barge/
- a working Kubernetes (K8s) cluster (Minikube is a good start)
- a working
kubectl
connected to the K8s cluster - one folder (/ocean/operator-service/), in which we will download the following:
- one folder (/ocean/operator-engine/), in which we will download the following:
Customize your Operator Service deployment
The following resources need attention:
- postgres-configmap.yaml - contains secrets for the postgresql deployment
- deployment.yaml - ALGO_POD_TIMEOUT = Allowed time for a algorithm to run. If it exceeded this value (in minutes), it's going to get killed
Customize your Operator Engine deployment
The following resources need attention:
- operator.yaml - contains configuration variabiles
- ACCOUNT_JSON , ACCOUNT_PASSWORD = Defines the account that is going to be used when publishing results back to OceanProtocol
- AWS_ACCESS_KEY_ID, AWS_ACCESS_KEY_ID , AWS_REGION = S3 credidentials for the logs and output buckets
- AWS_BUCKET_OUTPUT = Bucket that will hold the output data (algorithm logs & algorithm output)
- AWS_BUCKET_ADMINLOGS = Bucket that will hold the admin logs (logs from pod-configure & pod-publish)
- STORAGE_CLASS = Storage class to use (see next section)
Storage class
For minikube, you can use 'standard' class.
For AWS , please make sure that your class allocates volumes in the same region and zone in which you are running your pods.
We created our own 'standard' class in AWS:
kubectl get storageclass standard -o yaml
allowedTopologies:
- matchLabelExpressions:
- key: failure-domain.beta.kubernetes.io/zone
values:
- us-east-1a
apiVersion: storage.k8s.io/v1
kind: StorageClass
parameters:
fsType: ext4
type: gp2
provisioner: kubernetes.io/aws-ebs
reclaimPolicy: Delete
volumeBindingMode: Immediate
Or we can use this for minikube:
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
name: standard
provisioner: docker.io/hostpath
reclaimPolicy: Retain
For more information, please visit https://kubernetes.io/docs/concepts/storage/storage-classes/
Create namespaces
kubectl create ns ocean-operator
kubectl create ns ocean-compute
Deploy Operator Service
kubectl config set-context --current --namespace ocean-operator
kubectl create -f /ocean/operator-service/postgres-configmap.yaml
kubectl create -f /ocean/operator-service/postgres-storage.yaml
kubectl create -f /ocean/operator-service/postgres-deployment.yaml
kubectl create -f /ocean/operator-service/postgresql-service.yaml
kubectl apply -f /ocean/operator-service/deployment.yaml
kubectl apply -f /ocean/operator-service/role_binding.yaml
kubectl apply -f /ocean/operator-service/service_account.yaml
Deploy Operator Engine
kubectl config set-context --current --namespace ocean-compute
kubectl apply -f /ocean/operator-engine/sa.yml
kubectl apply -f /ocean/operator-engine/binding.yml
kubectl apply -f /ocean/operator-engine/operator.yml
kubectl apply -f /ocean/operator-engine/computejob-crd.yaml
kubectl apply -f /ocean/operator-engine/workflow-crd.yaml
kubectl create -f /ocean/operator-service/postgres-configmap.yaml
Expose Operator - Service
kubectl expose deployment operator-api --namespace=ocean-operator --port=8050
Run a port forward or create your ingress service (not covered here):
kubectl -n ocean-operator port-forward svc/operator-api 8050
Initialize database
If your cluster is running on example.com:
curl -X POST "http://example.com:8050/api/v1/operator/pgsqlinit" -H "accept: application/json"
Update Brizo
Update Brizo by adding or updating the OPERATOR_SERVICE_URL
env in /ocean/barge/compose-files/brizo.yaml
OPERATOR_SERVICE_URL: http://example.com:8050/
## Restart Barge with updated Brizo configuration