What's the relationship of kubernetes API server and kube-apiserver?

General Discussions
1

When I read doc of kube-apiserver:

The API server is a component of the Kubernetes control plane that exposes the Kubernetes API. The API server is the front end for the Kubernetes control plane.

The main implementation of a Kubernetes API server is kube-apiserver. kube-apiserver is designed to scale horizontally—that is, it scales by deploying more instances. You can run several instances of kube-apiserver and balance traffic between those instances

So the API server is not equals to kube-apiserver.
what’s the relathionship of them? is it parallel to OSI model and TCP/IP protocol? the standard and implementation?

question2:
by the way , the api in the diagram, is it mean the kubernetes API server or kube-apiserver?

image
image1252×585 70.6 KB

question3: there are severalapi instances, is it mean it is scale horizontally?

2

This means the abstract idea of a server which has the right REST semantics (consistent object model, watch, etc). Usually it includes the Kubernetes API types and schema, but there is a concept of a generic API server, which has the semantics but just a barebones schema, which can be used for other things (e.g. running your own system with your own types and schema).

This is a specific program which implements the Kubernetes API semantics and schema.

From the doc you quoted: “kube-apiserver is designed to scale horizontally—that is, it scales by deploying more instances.”