This is the source code for the Cloud Memorystore for Redis based leaderboard application that accompanies the tutorial: Using Cloud Memorystore for Redis as a game leaderboard. The application is built using ASP.NET Core and consists of a simple Web API for posting and retrieves scores. The Web API is implemented using Cloud Memorystore for Redis, which is a fully managed Redis service available through Google Cloud Platform (GCP).
These are the two key methods that make up the Web API:
Task<JsonResult> Post([FromBody] ScoreModel model)
Task<JsonResult> RetrieveScores(string centerKey, int offset, int numScores)This Web API is called from a simple Vue.js game to post scores and retrieve the leaderboard.
There are two high-level steps involved in deploying the application:
- Upload Docker image to a container registry
- Deploy leaderboard application to a Kubernetes cluster
To upload the Docker image for the leaderboard application to your private registry with GCR, run build.sh. This script also builds the ASP.NET application using Cloud Build and then configures the image for deployment.
For more information about GCR, please see Container Registry Documentation.
For more information about Cloud Build, please see Cloud Build Documentation.
To set up a GKE cluster, please consult the Google Kubernetes Engine Documentation.
NOTE
Within GKE, Alias IP addresses need to be turned on to allow your application pods to reach the Cloud Memorystore for Redis instances in the same VPC network. Please see Creating VPC-native clusters using alias IP addresses for more details.
Once you have a cluster ready and have kubectl configured from your local client, you can deploy the application and service using the appdeploy.yaml file (in the leaderboardapp folder). First, open appdeploy.yaml in your favorite editor and replace YOUR_PROJECT_ID with your project id.
image: "gcr.io/[YOUR_PROJECT_ID]/leaderboardapp"And then deploy using kubectl:
kubectl create -f appdeploy.yaml
kubectl create -f gamingservice.yamlOnce the application and service have been deployed, the ingress component needs to be configured. With GKE, ingress is provided by the GCP HTTP(s) load balancer as described in HTTP(s) load balancing with Ingress. To deploy ingress, use kubectl again:
kubectl create -f appingress.yamlThe ingress component takes anywhere from 2-20 minutes to become available. Once it's available, you can navigate to the front end IP address associated with HTTP(s) load balancer.