Linked item names, (esp. "Data" node) and speed analysis of extra nodes

Hi, I tend to use the "Data" node a lot in order to rename data that's closer to the newer set of nodes so I understand what the information is. Sometimes, though, I have realised that I need to change the name of the initial data, so all the linked data noded with a custom name have to be renamed by hand. Is there a way to automate the renamig process, or perhaps a suggestion for an alternate workflow that has some sort of linked names.

My other question is whether the speed of the calculation would be hindered by all the extra data nodes that I use. It seems thus far that that each node copies information from previous one so I imagine if i have a curve node with 1000 curves, and 6 data nodes linking to that curve node that perhaps, it slows down the calculation?

In general, how do you optimise very large documents with numerous node groups and keep naming things in a way that is easy to change later down the process? Could there be some sort of custom node with linked name? Is a cluster the best option for a simple node, or would that slow down things?

Thanks :)