Quickstart with Docker Desktop

This guide provides instructions to deploy a single-server TigerGraph instance to a local Kubernetes cluster via Docker Desktop.

Before you begin

Install Docker Desktop on your computer.
Install the Kubernetes CLI kubectl.
Enable the Docker Desktop Kubernetes engine. For more information, see Docker official documentation.
Verify that you have sufficient CPU, RAM and DISK for your TigerGraph deployment.

Procedure

1. Download Kubernetes manifest

Download tigergraph-docker.yaml. This is the file that describes your Kubernetes deployment.

2. Create deployment

Make sure that you have configured kubectl to use the correct context. In this case, the name of the context should be docker-desktop. Then run kubectl apply to create the deployment. Replace <path_to_tigergraph.yaml> with the path to the tigergraph-docker.yaml file you downloaded in the step before.

$ kubectl apply -f <path_to_tigergraph.yaml>

3. Verify deployment

Verify that the pods were created successfully:

$ kubectl get pods -l app=tigergraph -w

4. Test deployment

The deployment maps port 9000 of the pod to port 30090 of localhost. We can test that the REST API is up and running by making a curl call to port 30090:

$ curl http://localhost:30090/echo
{"error":false, "message":"Hello GSQL"}%

Port 14240 of the pod is mapped to port 30240 of localhost. Visit http://localhost:30240 to access GraphStudio.

Cleanup

Run the following command to delete all cluster resources:

$ kubectl delete -f <path_to_tigergraph.yaml> &&  kubectl delete pvc -l app=tigergraph