I created an analytic on a small graph, how easy is it to scale my analytic to large graph?
At the python layer, cuGraph leverages dask to distribute the work/data accross multiple GPUs. For that transition, one has to setup the dask cluster and the MG API almost matches the SG one. Checkout more here: Multi-GPU with cuGraph — cugraph 23.02.00 documentation
1 Like
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.