How to solve the problem of single-machine memory explosion caused by large graphics?

[ghost]

Is your feature request related to a problem? Please describe.
A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]

Describe the solution you'd like
A clear and concise description of what you want to happen.

Describe alternatives you've considered
A clear and concise description of any alternative solutions or features you've considered.

Additional context
Add any other context or screenshots about the feature request here.

[sighingnow]

Hi @Illcat I have move the ticket to the discussion category.

GraphScope distributes large graphs to many fragments and those fragments can be placed on a set of machines in a large-scale cluster. The computation engines in GraphScope scales pretty well on distributed large graphs.