IP Filtering, Deep Dive

In the other topic, @Vadim mentioned that he has a few nodes in the same /24 subnet and the total amount of data of about 25TB.

So, it would seem that multiple nodes in the same /24 are not completely aggregated into one big node and running multiple nodes results in more data.
I would like clarification on that. Is it only for nodes with different IPs (same /24) or would that apply to nodes with the same IP?
I have free space on my server, could spin up a few more VMs…