A county-fair volunteer-crew worked-with graph grew from a classroom clique to tens of thousands of edges. The team wants the same shortest-path and importance queries they already trust. How should they design the runtime experiment?
Select an answer to reveal the explanation.
Short Explanation
The same shortest-path and importance queries, now on tens of thousands of edges. Time NetworkX on CPU versus cuGraph on GPU. The graph theory does not change. Use the GPU treatment when the graph is huge and a GPU is there.
Full Explanation
NetworkX and cuGraph implement the same family of graph questions; the experimental factor is scale and runtime, not a new theory. On a modest clique, CPU NetworkX is enough. When edges grow into the tens of thousands and a GPU is available, a timed cuGraph treatment on the same queries is the run to schedule. Hand-written kernels and multimodal image graphs are the wrong treatments.