I am running a Kubernates Cluster in GCP. Each one of my clients/customer will need their own Kubernates Cluster.

Each Cluster/client needs to have their own database. What is the best way to go about this? I would like each cluster to have a MongoDB service so every pod in the cluster can communicate with the DB, but what is the correct way to do this? Create a pod running MongoDb with persistent storage on the node?

Create a replica set of mongoDB pods inside the cluster?

Have a single large MongoDB that each client's cluster can communicate with?