Function Differences with torch.distributed.init_process_group

torch.distributed.init_process_group

torch.distributed.init_process_group(
    backend,
    init_method=None,
    timeout=datetime.timedelta(0, 1800),
    world_size=-1,
    rank=-1,
    store=None,
    group_name=''
)

For more information, see torch.distributed.init_process_group.

mindspore.communication.init

mindspore.communication.init(backend_name=None)

For more information, see mindspore.communication.init.

Differences

PyTorch: This interface supports three kinds of collective communications: MPI, Gloo, and NCCL. It initializes backend and also provides configuration, such as world_size, rank, timeout, etc.

MindSpore：This interface currently supports three kinds of collective communication: HCCL, NCCL, and MCCL. The configuration of world_size, rank and timeout is not set in this interface. The corresponding environment variables need to be set before calling this interface.

Class	Sub-class	PyTorch	MindSpore	Difference
Param	Param 1	backend	backend_name	The functionalities are the same, but have different names and support different communication backend
	Param 2	init_method	-	PyTorch: the initialization method. MindSpore does not have this param
	Param 3	timeout	-	PyTorch: the timeout value. MindSpore does not have this param, and should set the corresponding environment variables before calling this interface
	Param 4	world_size	-	PyTorch: the world size of communication group. MindSpore does not have this param, and should set the corresponding environment variables before calling this interface
	Param 5	rank	-	PyTorch: the current rank. MindSpore does not have this param, and should set the corresponding environment variables before calling this interface
	Param 6	store	-	PyTorch: the flag of storing key/value. MindSpore does not have this param
	Param 7	group_name	-	PyTorch: the group name. MindSpore does not have this param