how do you roughly evaluate hardware resources when deploying a Nebula cluster?
For example, to store 10 million points and 1 trillion edges, assuming each point has only one Tag, how many nodes should be used at least, and what should be the minimum configuration of each node.
