There are two levels of load balancing in Kubernetes. The upper level is provided by the infrastructure of your cloud provider, which gets provisioned when you create a LoadBalancer Service. These load balancers are responsible for distributing traffic from end users to Kubernetes nodes. The lower-level load balancing mechanism is provided by Kubernetes itself. It distributes traffic among pods of the same service, after user requests have reached Kubernetes nodes through the upper-level LoadBalancer service. If there is no pod running on the node for the requested Kubernetes application, the node forwards the traffic to a node where there is a running pod for the service.