| | import functools |
| | import pickle |
| | import torch |
| | import torch.distributed as dist |
| |
|
| | import logging |
| | logger = logging.getLogger(__name__) |
| |
|
| | |
| |
|
| | def is_dist_avail_and_initialized(): |
| | if not dist.is_available(): |
| | return False |
| | if not dist.is_initialized(): |
| | return False |
| | return True |
| |
|
| |
|
| | def get_rank(): |
| | """ |
| | Get the rank of the current process. |
| | """ |
| | if not is_dist_avail_and_initialized(): |
| | return 0 |
| | return dist.get_rank() |
| |
|
| |
|
| | def get_world_size(): |
| | """ |
| | Get the size of the world. |
| | """ |
| | if not is_dist_avail_and_initialized(): |
| | return 1 |
| | return dist.get_world_size() |
| |
|
| |
|
| | def is_master_proc(num_gpus=8): |
| | """ |
| | Determines if the current process is the master process on each node. |
| | """ |
| | if is_dist_avail_and_initialized(): |
| | return dist.get_rank() % num_gpus == 0 |
| | else: |
| | return True |
| |
|
| |
|
| | def is_root_proc(): |
| | """ |
| | Determines if the current process is the root process. |
| | """ |
| | if is_dist_avail_and_initialized(): |
| | return dist.get_rank() == 0 |
| | else: |
| | return True |
| |
|
| |
|
| | |
| |
|
| | def _serialize_to_tensor(data, group, max_size=1024): |
| | """ |
| | Serialize the tensor to ByteTensor. Note that only `gloo` and `nccl` |
| | backend is supported. |
| | Args: |
| | data (data): data to be serialized. |
| | group (group): pytorch dist group. |
| | Returns: |
| | tensor (ByteTensor): tensor that serialized. |
| | """ |
| | backend = dist.get_backend(group) |
| | assert backend in ["gloo", "nccl"] |
| | device = torch.device("cpu" if backend == "gloo" else "cuda") |
| |
|
| | buffer = pickle.dumps(data) |
| | if len(buffer) > max_size ** 3: |
| | logger.warning( |
| | "Rank {} trying to all-gather {:.2f} GB of data on device {}".format( |
| | get_rank(), len(buffer) / (max_size ** 3), device |
| | ) |
| | ) |
| | storage = torch.ByteStorage.from_buffer(buffer) |
| | tensor = torch.ByteTensor(storage).to(device=device) |
| | return tensor |
| |
|
| |
|
| | def _pad_to_largest_tensor(tensor, group): |
| | """ |
| | Padding all the tensors from different GPUs to the largest ones. |
| | Args: |
| | tensor (tensor): tensor to pad. |
| | group (group): pytorch dist group. |
| | Returns: |
| | list[int]: size of the tensor, on each rank |
| | Tensor: padded tensor that has the max size |
| | """ |
| | world_size = dist.get_world_size(group=group) |
| | assert ( |
| | world_size >= 1 |
| | ), "comm.gather/all_gather must be called from ranks within the given group!" |
| | local_size = torch.tensor( |
| | [tensor.numel()], dtype=torch.int64, device=tensor.device |
| | ) |
| | size_list = [ |
| | torch.zeros([1], dtype=torch.int64, device=tensor.device) |
| | for _ in range(world_size) |
| | ] |
| | dist.all_gather(size_list, local_size, group=group) |
| | size_list = [int(size.item()) for size in size_list] |
| |
|
| | max_size = max(size_list) |
| |
|
| | |
| | |
| | if local_size != max_size: |
| | padding = torch.zeros( |
| | (max_size - local_size,), dtype=torch.uint8, device=tensor.device |
| | ) |
| | tensor = torch.cat((tensor, padding), dim=0) |
| | return size_list, tensor |
| |
|
| |
|
| | def broadcast(object): |
| | if isinstance(object, torch.Tensor): |
| | dist.broadcast(tensor=object, src=0) |
| | else: |
| | sync_tensor = torch.Tensor([object]).cuda() |
| | dist.broadcast(tensor=sync_tensor, src=0) |
| | object = sync_tensor[0].item() |
| | return object |
| |
|
| |
|
| | def all_gather(tensors): |
| | """ |
| | All gathers the provided tensors from all processes across machines. |
| | Args: |
| | tensors (list): tensors to perform all gather across all processes in |
| | all machines. |
| | """ |
| | gather_list = [] |
| | output_tensor = [] |
| | world_size = dist.get_world_size() |
| | for tensor in tensors: |
| | tensor_placeholder = [ |
| | torch.ones_like(tensor) for _ in range(world_size) |
| | ] |
| | dist.all_gather(tensor_placeholder, tensor, async_op=False) |
| | gather_list.append(tensor_placeholder) |
| | for gathered_tensor in gather_list: |
| | output_tensor.append(torch.cat(gathered_tensor, dim=0)) |
| | return output_tensor |
| |
|
| |
|
| | def all_reduce(tensors, average=True): |
| | """ |
| | All reduce the provided tensors from all processes across machines. |
| | Args: |
| | tensors (list): tensors to perform all reduce across all processes in |
| | all machines. |
| | average (bool): scales the reduced tensor by the number of overall |
| | processes across all machines. |
| | """ |
| | for tensor in tensors: |
| | dist.all_reduce(tensor, async_op=False) |
| | if average: |
| | world_size = dist.get_world_size() |
| | for tensor in tensors: |
| | tensor.mul_(1.0 / world_size) |
| | return tensors |
| |
|
| |
|
| | @functools.lru_cache() |
| | def _get_global_gloo_group(): |
| | """ |
| | Return a process group based on gloo backend, containing all the ranks |
| | The result is cached. |
| | Returns: |
| | (group): pytorch dist group. |
| | """ |
| | if dist.get_backend() == "nccl": |
| | return dist.new_group(backend="gloo") |
| | else: |
| | return dist.group.WORLD |
| |
|
| |
|
| | def all_gather_unaligned(data, group=None): |
| | """ |
| | Run all_gather on arbitrary picklable data (not necessarily tensors). |
| | |
| | Args: |
| | data: any picklable object |
| | group: a torch process group. By default, will use a group which |
| | contains all ranks on gloo backend. |
| | |
| | Returns: |
| | list[data]: list of data gathered from each rank |
| | """ |
| | if get_world_size() == 1: |
| | return [data] |
| | if group is None: |
| | group = _get_global_gloo_group() |
| | if dist.get_world_size(group) == 1: |
| | return [data] |
| |
|
| | tensor = _serialize_to_tensor(data, group) |
| |
|
| | size_list, tensor = _pad_to_largest_tensor(tensor, group) |
| | max_size = max(size_list) |
| |
|
| | |
| | tensor_list = [ |
| | torch.empty((max_size,), dtype=torch.uint8, device=tensor.device) |
| | for _ in size_list |
| | ] |
| | dist.all_gather(tensor_list, tensor, group=group) |
| |
|
| | data_list = [] |
| | for size, tensor in zip(size_list, tensor_list): |
| | buffer = tensor.cpu().numpy().tobytes()[:size] |
| | data_list.append(pickle.loads(buffer)) |
| |
|
| | return data_list |