What's the difference between the datastores and how do they operate?

Hi guys, just wanna ask what's the difference between them and how to know which deployment is better?

I know that the virtual storage is like a virtual pool by combining 2 physical storage or more as 1. For an example, the setup is using 2 nodes, node A and node B. If node A is down, node B can still run the VM as long it has the capacity to run the VM that initially ran on node A.

So does the other datastore operates in the same way as the virtual storage but different approach or they are completely different?

If possible I would like a clear comparison between the datastores regarding their types and how do they operate differently. I can just google it but i feel like maybe in Sangfor it may use a different apprach, so that's why I'm asking.