[RuntimeError: Connection reset by peer] When scaling up training jobs

I am facing a similar problem as the one posted by @g-karthik in https://github.com/microsoft/DeepSpeed/issues/570#issuecomment-750744107.

When I use 40 nodes with 10 gpus on each node (400 jobs), the training works well. But when I scale up the training to 40 or more nodes, deepspeed.initialize() fails with:

Traceback (most recent call last):  File "/home/hanwentao/work/enc-dec-pretrain/Megatron-LM/pretrain_enc_dec.py", line 947, in <module>
    main()  
File "/home/hanwentao/work/enc-dec-pretrain/Megatron-LM/pretrain_enc_dec.py", line 769, in main
    initialize_distributed(args)  
File "/home/hanwentao/work/enc-dec-pretrain/Megatron-LM/pretrain_enc_dec.py", line 703, in initialize_distributed
    deepspeed.init_distributed(distributed_port=29501)  
File "/home/hanwentao/.local/lib/python3.8/site-packages/deepspeed-0.3.11+4f1d827-py3.8.egg/deepspeed/utils/distributed.py", line 49, in init_distributed                                          
    torch.distributed.init_process_group(backend=dist_backend,  
File "/home/hanwentao/.local/lib/python3.8/site-packages/torch/distributed/distributed_c10d.py", line 455, in init_process_group
    barrier()  
File "/home/hanwentao/.local/lib/python3.8/site-packages/torch/distributed/distributed_c10d.py", line 1960, in barrier
    work = _default_pg.barrier()
RuntimeError: Connection reset by peer

I used the deepspeed version at the master branch. I ran my script with mpirun, just as described in https://www.deepspeed.ai/getting-started/#mpi-and-azureml-compatibility.

Any ideas on what's going on?

Start with the reported deepspeed.init_distributed call and the torch.distributed.init_process_group/barrier frames in the traceback. Compare the mpirun setup that works at 40 nodes with the larger-scale run, focusing on distributed initialization and the peer-reset failure. Done means identifying a reproducible cause or a confirmed configuration fix for scaling beyond 40 nodes.