Closed parthosa closed 2 weeks ago
Currently for cluster information deduction, we calculate the number of nodes correctly but do not track the number of executors per node.
This can generate wrong GPU cluster recommendations because there can be multiple executors per node. We should track number of executors per node.
Note this is basically a subtask of fixing https://github.com/NVIDIA/spark-rapids-tools/issues/1109
Currently for cluster information deduction, we calculate the number of nodes correctly but do not track the number of executors per node.
This can generate wrong GPU cluster recommendations because there can be multiple executors per node. We should track number of executors per node.