Accelerator abstraction

deepspeed/accelerator/__init__.py

@@ -0,0 +1 @@
+from .abstract_accelerator import DeepSpeedAccelerator

deepspeed/accelerator/abstract_accelerator.py

@@ -0,0 +1,161 @@
+import abc
+from abc import ABC
+
+
+class DeepSpeedAccelerator(ABC):
+    def __init__(self):
+        self._name = None
+        self._communication_backend_name = None
+        self.BFloat16Tensor = None
+        self.ByteTensor = None
+        self.DoubleTensor = None
+        self.FloatTensor = None
+        self.HalfTensor = None
+        self.IntTensor = None
+        self.LongTensor = None
+
+    # Device APIs
+    @abc.abstractmethod
+    def device_name(self, device_index):
+        ...
+
+    @abc.abstractmethod
+    def device(self, device_index):
+        ...
+
+    @abc.abstractmethod
+    def set_device(self):
+        ...
+
+    @abc.abstractmethod
+    def current_device(self):
+        ...
+
+    @abc.abstractmethod
+    def device_count(self):
+        ...
+
+    @abc.abstractmethod
+    def synchronize(self, device_index=None):
+        ...
+
+    # RNG APIs
+    @abc.abstractmethod
+    def set_rng_state(self, new_state, device_index=None):
+        ...
+
+    @abc.abstractmethod
+    def get_rng_state(self, device_index=None):
+        ...
+
+    @abc.abstractmethod
+    def manual_seed(self, seed):
+        ...
+
+    @abc.abstractmethod
+    def manual_seed_all(self, seed):
+        ...
+
+    @abc.abstractmethod
+    def initial_seed(self):
+        ...
+
+    @abc.abstractmethod
+    def default_generator(self, device_index):
+        ...
+
+    # Streams/Events
+    @abc.abstractmethod
+    def Stream(self, device_index=None, priority=0, **kwargs):
+        ...
+
+    @abc.abstractmethod
+    def StreamContext(self, stream):
+        ...
+
+    @abc.abstractmethod
+    def current_stream(self, device_index=None):
+        ...
+
+    @abc.abstractmethod
+    def default_stream(self, device_index=None):
+        ...
+
+    @abc.abstractmethod
+    def Event(self, **kwargs):
+        ...
+
+    # Memory management
+    @abc.abstractmethod
+    def empty_cache(self):
+        ...
+
+    @abc.abstractmethod
+    def memory_allocated(self, device_index=None):
+        ...
+
+    @abc.abstractmethod
+    def max_memory_allocated(self, device_index=None):
+        ...
+
+    @abc.abstractmethod
+    def reset_max_memory_allocated(self, device_index=None):
+        ...
+
+    @abc.abstractmethod
+    def reset_max_memory_cached(self, device_index=None):
+        ...
+
+    @abc.abstractmethod
+    def memory_stats(self, device_index=None):
+        ...
+
+    @abc.abstractmethod
+    def reset_peak_memory_stats(self, device_index=None):
+        ...
+
+    @abc.abstractmethod
+    def memory_reserved(self, device_index=None):
+        ...
+
+    @abc.abstractmethod
+    def max_memory_reserved(self, device_index=None):
+        ...
+
+    @abc.abstractmethod
+    def total_memory(self, device_index=None):
+        ...
+
+    # Data types
+    @abc.abstractmethod
+    def is_bf16_supported(self):
+        ...
+
+    @abc.abstractmethod
+    def is_fp16_supported(self):
+        ...
+
+    # Misc
+    @abc.abstractmethod
+    def is_available(self):
+        ...
+
+    @abc.abstractmethod
+    def range_push(self, msg):
+        ...
+
+    @abc.abstractmethod
+    def range_pop(self, msg):
+        ...
+
+    @abc.abstractmethod
+    def lazy_call(self, callback):
+        ...
+
+    @abc.abstractmethod
+    def name(self):
+        ...
+
+    @abc.abstractmethod
+    def communication_backend_name(self):
+        ...

deepspeed/accelerator/cuda_accelerator.py

@@ -0,0 +1,139 @@
+from deepspeed.accelerator.abstract_accelerator import DeepSpeedAccelerator
+import torch.cuda
+
+
+class CUDA_Accelerator(DeepSpeedAccelerator):
+    def __init__(self):
+        self._name = 'cuda'
+        self._communication_backend_name = 'nccl'
+        self.DoubleTensor = torch.cuda.DoubleTensor
+        self.LongTensor = torch.cuda.LongTensor
+        self.FloatTensor = torch.cuda.FloatTensor
+        self.BFloat16Tensor = torch.cuda.BFloat16Tensor
+        self.HalfTensor = torch.cuda.HalfTensor
+        self.IntTensor = torch.cuda.IntTensor
+        self.ByteTensor = torch.cuda.ByteTensor
+
+    # Device APIs
+    def device_name(self, device_index=None):
+        idx = torch.cuda.current_device() if device_index is None else device_index
+        return f'cuda:{idx}'
+
+    def device(self, device_index=None):
+        return torch.cuda.device(device_index)
+
+    def set_device(self, device_index):
+        torch.cuda.set_device(device_index)
+
+    def current_device(self):
+        return torch.cuda.current_device()
+
+    def device_count(self):
+        return torch.cuda.device_count()
+
+    def synchronize(self, device_index=None):
+        return torch.cuda.synchronize(device_index)
+
+    # RNG APIs
+    def set_rng_state(self, new_state, device_index=None):
+        if device_index is None:
+            return torch.cuda.set_rng_state(new_state)
+
+        return torch.cuda.set_rng_state(new_state, device_index)
+
+    def get_rng_state(self, device_index=None):
+        if device_index is None:
+            return torch.cuda.get_rng_state()
+
+        return torch.cuda.get_rng_state(device_index)
+
+    def manual_seed(self, seed):
+        return torch.cuda.manual_seed(seed)
+
+    def manual_seed_all(self, seed):
+        return torch.cuda.manual_seed_all(seed)
+
+    def initial_seed(self, seed):
+        return torch.cuda.initial_seed(seed)
+
+    def default_generator(self, device_index):
+        return torch.cuda.default_generators[device_index]
+
+    # Streams/Events
+    def Stream(self, device_index=None, priority=0, **kwargs):
+        return torch.cuda.Stream(device_index, priority, **kwargs)
+
+    def StreamContext(self, stream):
+        return torch.cuda.StreamContext(stream)
+
+    def current_stream(self, device_index=None):
+        return torch.cuda.current_stream(device_index)
+
+    def default_stream(self, device_index=None):
+        return torch.cuda.default_stream(device_index)
+
+    def Event(self, **kwargs):
+        return torch.cuda.Event(**kwargs)
+
+    # Memory management
+    def empty_cache(self):
+        return torch.cuda.empty_cache()
+
+    def memory_allocated(self, device_index=None):
+        return torch.cuda.memory_allocated(device_index)
+
+    def max_memory_allocated(self, device_index=None):
+        return torch.cuda.max_memory_allocated(device_index)
+
+    def reset_max_memory_allocated(self, device_index=None):
+        return torch.cuda.reset_max_memory_allocated(device_index)
+
+    def reset_max_memory_cached(self, device_index=None):
+        return torch.cuda.reset_max_memory_cached(device_index)
+
+    def memory_stats(self, device_index=None):
+        if hasattr(torch.cuda, 'memory_stats'):
+            return torch.cuda.memory_stats(device_index)
+
+    def reset_peak_memory_stats(self, device_index=None):
+        if hasattr(torch.cuda, 'reset_peak_memory_stats'):
+            return torch.cuda.reset_peak_memory_stats(device_index)
+
+    def memory_reserved(self, device_index=None):
+        if hasattr(torch.cuda, 'memory_reserved'):
+            return torch.cuda.memory_reserved(device_index)
+
+    def max_memory_reserved(self, device_index=None):
+        if hasattr(torch.cuda, 'max_memory_reserved'):
+            return torch.cuda.max_memory_reserved(device_index)
+
+    def total_memory(self, device_index=None):
+        return torch.cuda.get_device_properties(device_index).total_memory
+
+    # Data types
+    def is_bf16_supported(self):
+        return torch.cuda.is_bf16_supported()
+
+    def is_fp16_supported(self):
+        return torch.cuda.is_fp16_supported()
+
+    # Misc
+    def is_available(self):
+        return torch.cuda.is_available()
+
+    def range_push(self, msg):
+        if hasattr(torch.cuda.nvtx, 'range_push'):
+            return torch.cuda.nvtx.range_push(msg)
+
+    def range_pop(self, msg):
+        if hasattr(torch.cuda.nvtx, 'range_pop'):
+            return torch.cuda.nvtx.range_pop(msg)
+
+    def lazy_call(self, callback):
+        return torch.cuda._lazy_call(callback)
+
+    def name(self):
+        return self._name
+
+    def communication_backend_name(self):
+        return self._communication_backend_name

deepspeed/accelerator/real_accelerator.py

@@ -0,0 +1,71 @@
+from .abstract_accelerator import DeepSpeedAccelerator
+
+ds_accelerator = None
+
+
+def _validate_accelerator(accel_obj):
+    assert isinstance(accel_obj, DeepSpeedAccelerator), \
+        f'{accel_obj.__class__.__name__} accelerator is not subclass of DeepSpeedAccelerator'
+
+    assert accel_obj.is_available(), \
+        f'{accel_obj.__class__.__name__} accelerator fails is_available() test'
+
+
+def get_accelerator():
+    global ds_accelerator
+    if ds_accelerator is None:
+        from deepspeed.accelerator.cuda_accelerator import CUDA_Accelerator
+        ds_accelerator = CUDA_Accelerator()
+        _validate_accelerator(ds_accelerator)
+    return ds_accelerator
+
+
+def set_accelerator(accel_obj):
+    global ds_accelerator
+    _validate_accelerator(accel_obj)
+    ds_accelerator = accel_obj
+
+
+'''
+-----------[code] test_get.py -----------
+from deepspeed.accelerator.real_accelerator import get_accelerator
+my_accelerator = get_accelerator()
+print(f'{my_accelerator.name=}')
+print(f'{my_accelerator.communication_backend=}')
+print(f'{my_accelerator.HalfTensor().device=}')
+print(f'{my_accelerator.total_memory()=}')
+-----------[code] test_get.py -----------
+
+---[output] python test_get.py---------
+my_accelerator.name()='cuda'
+my_accelerator.communication_backend='nccl'
+my_accelerator.HalfTensor().device=device(type='cuda', index=0)
+my_accelerator.total_memory()=34089730048
+---[output] python test_get.py---------
+
+**************************************************************************
+-----------[code] test_set.py -----------
+from deepspeed.accelerator.cuda_accelerator import CUDA_Accelerator
+cu_accel = CUDA_Accelerator()
+print(f'{id(cu_accel)=}')
+from deepspeed.accelerator.real_accelerator import set_accelerator, get_accelerator
+set_accelerator(cu_accel)
+
+my_accelerator = get_accelerator()
+print(f'{id(my_accelerator)=}')
+print(f'{my_accelerator.name=}')
+print(f'{my_accelerator.communication_backend=}')
+print(f'{my_accelerator.HalfTensor().device=}')
+print(f'{my_accelerator.total_memory()=}')
+-----------[code] test_set.py -----------
+
+
+---[output] python test_set.py---------
+id(cu_accel)=139648165478304
+my_accelerator=<deepspeed.accelerator.cuda_accelerator.CUDA_Accelerator object at 0x7f025f4bffa0>
+my_accelerator.name='cuda'
+my_accelerator.communication_backend='nccl'
+my_accelerator.HalfTensor().device=device(type='cuda', index=0)
+my_accelerator.total_memory()=34089730048
+---[output] python test_set.py---------
+'''