Here’s the situation:
I created a series of global vertexes and edges using the CLI, defined a schema, and then populated a graph (called my_graph) with data. However, after playing with the graph, I noticed that there were several data errors (introduced by messy data - not TG related).
Using GraphStudio, I deleted my_graph (using the little trashcan icon) and then I created a second graph called my_new_graph. During creation, I chose the set of vertexes and edges that I had previously defined, but did not any data.
Since I have not yet added data, I would expect my_new_graph to be empty. However, when I navigate to “my_new_graph” and then click “Pick vertices” under the “explore graph tab”, a number of vertexes from the previous graph appear.
My question: What is TigerGraph’s data model? From this, it seems like when data that is loaded it is then stored somewhere as nodes and edges, which is independent of any specific graph. Later, when you build a graph, you are just relating these existing data stores to each other. Is this accurate? How can I clear out the underlying data stores? I want to maintain the schema I defined, but I do not want any data associated with it from the beginning; I want to be able to populate new graphs with a fresh set data.