[perf] ShardedDDP - small memory use reduction - minor speedup (#366)

* minor

* minor

fairscale/nn/data_parallel/sharded_ddp.py

@@ -142,6 +142,7 @@ class ShardedDataParallel(nn.Module):
 
         self.buckets: Dict[OSS, Dict[torch.device, List[Bucket]]] = {o: {} for o in self.sharded_optimizers}
         self._should_bucket_grad: List[bool] = []
+        self._bucket_list: Optional[List[Bucket]] = None
         self._setup_bucket_strategy()
 
         # - setup backward hooks which will be called by Torch's autograd in due time
@@ -155,6 +156,8 @@ class ShardedDataParallel(nn.Module):
         if sync_models_at_startup:
             self._sync_params_and_buffers()
 
+        self._clear_counters()
+
     def forward(self, *inputs: Any, **kwargs: Any) -> Any:
         """
         Module forward pass, handles any DDP-specific work in the background. Primes the
@@ -256,9 +259,10 @@ class ShardedDataParallel(nn.Module):
             self._grad_to_be_reduced = [True for _ in self._grad_to_be_reduced]
             self._reduced_grads = {o: 0 for o in self.sharded_optimizers}
 
-            for optimizer in self.buckets.keys():
-                for device in self.buckets[optimizer].keys():
-                    for bucket in self.buckets[optimizer][device]:
+            if self.use_buckets:
+                assert self._bucket_list is not None
+
+                for bucket in self._bucket_list:
                     assert bucket.sent, (
                         "A bucket failed to be sent, probably unused parameters."
                         + "Either remove the unused parameter or de-activate ShardedDDP buckets -set reduce_buffer_size to 0-"
@@ -374,18 +378,15 @@ class ShardedDataParallel(nn.Module):
             def bucket_flush(*unused: Any) -> None:
                 handle = None
 
-                for bucket_optim in self.buckets.values():
-                    for bucket_rank in bucket_optim.values():
-                        for bucket in bucket_rank:
+                assert self._bucket_list is not None
+
+                for bucket in self._bucket_list:
                     if not bucket.sent:
                         # Reduce the bucket. Some parameters went unused and this bucket was not flushed
                         bucket.buffer.mul_(self.world_size_scaling)
                         bucket.sent = True
                         handle = dist.reduce(
-                                    tensor=bucket.buffer,
-                                    dst=bucket.destination,
-                                    group=self.process_group,
-                                    async_op=True,
+                            tensor=bucket.buffer, dst=bucket.destination, group=self.process_group, async_op=True,
                         )
 
                 # Only wait on the last handle
@@ -430,19 +431,19 @@ class ShardedDataParallel(nn.Module):
         if not self.use_buckets:
             return
 
-        # - Allocate one buffer per rank and per device to group the small parameters
-        for sharded_optimizer in self.sharded_optimizers:
-            for device, per_device in sharded_optimizer.per_device_params.items():
-                self.buckets[sharded_optimizer][device] = [
-                    Bucket(buffer=torch.zeros(self.buffer_max_size, dtype=per_device[0][0].dtype, device=device))
-                    for _ in per_device
-                ]
-
         # Devise the bucketing strategy
         for sharded_optimizer in self.sharded_optimizers:
             for device, per_rank_params in sharded_optimizer.per_device_params.items():
+                self.buckets[sharded_optimizer][device] = []
+
                 for dst_rank, params in enumerate(per_rank_params):
                     offset = 0
+
+                    self.buckets[sharded_optimizer][device].append(
+                        Bucket(
+                            buffer=torch.zeros(self.buffer_max_size, dtype=per_rank_params[0][0].dtype, device=device)
+                        )
+                    )
                     bucket = self.buckets[sharded_optimizer][device][dst_rank]
                     bucket.destination = dst_rank
 
@@ -473,3 +474,13 @@ class ShardedDataParallel(nn.Module):
                     bucket.buffer.resize_(offset)
                     if bucket.max_params_checked_in > 0:
                         self._reduced_grads_max[sharded_optimizer] += 1  # one reduce call per bucket
+
+        self._bucket_list = list(
+            chain(
+                *[
+                    self.buckets[sharded_optimizer][device]
+                    for sharded_optimizer in self.sharded_optimizers
+                    for device in sharded_optimizer.per_device_params.keys()
+                ]
+            )
+        )