You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This is my experiment environment.
python: 3.8.10
pytorch: 2.1.0
pytorch_geometric: 2.5.3
It works well if you divide the ogbn-products dataset into 4 partitions and perform distributed learning on 4 nodes. However, if you try to do distributed training with 8 nodes with 8 partitions, an error occurs.
RPC agent for mp_sampling_worker-2 encountered error when sending outgoing request #8 to mp_sampling_worker-12:
Coroutine task failed with error:
It may be a problem with the shared memory (/dev/shm), so I checked and found that the allocated memory was 56G, which I think is sufficient.
Additionally, there are no problems with communication between all nodes.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
I am running this example from github.
https://github.com/pyg-team/pytorch_geometric/tree/master/examples/distributed/pyg#option-a-manual-execution
This is my experiment environment.
python: 3.8.10
pytorch: 2.1.0
pytorch_geometric: 2.5.3
It works well if you divide the ogbn-products dataset into 4 partitions and perform distributed learning on 4 nodes.
However, if you try to do distributed training with 8 nodes with 8 partitions, an error occurs.
RPC agent for mp_sampling_worker-2 encountered error when sending outgoing request #8 to mp_sampling_worker-12:
Coroutine task failed with error:
It may be a problem with the shared memory (/dev/shm), so I checked and found that the allocated memory was 56G, which I think is sufficient.
Additionally, there are no problems with communication between all nodes.
Is there anything else I should check?
Beta Was this translation helpful? Give feedback.
All reactions