[data] Refactor all to all op implementations into a separate file (#26585)

python/ray/data/_internal/plan.py

@@ -194,6 +194,8 @@ class ExecutionPlan:
         Returns:
             The schema of the output dataset.
         """
+        from ray.data._internal.stage_impl import RandomizeBlocksStage
+
         if self._stages_after_snapshot:
             if fetch_if_missing:
                 if isinstance(self._stages_after_snapshot[-1], RandomizeBlocksStage):
@@ -396,6 +398,8 @@ class ExecutionPlan:
 
     def is_read_stage_equivalent(self) -> bool:
         """Return whether this plan can be executed as only a read stage."""
+        from ray.data._internal.stage_impl import RandomizeBlocksStage
+
         context = DatasetContext.get_current()
         remaining_stages = self._stages_after_snapshot
         if (
@@ -712,20 +716,6 @@ class AllToAllStage(Stage):
         return blocks, stage_info
 
 
-class RandomizeBlocksStage(AllToAllStage):
-    def __init__(self, seed: Optional[int]):
-        self._seed = seed
-
-        super().__init__("randomize_block_order", None, self.do_randomize)
-
-    def do_randomize(self, block_list, *_):
-        num_blocks = block_list.initial_num_blocks()
-        if num_blocks == 0:
-            return block_list, {}
-        randomized_block_list = block_list.randomize_block_order(self._seed)
-        return randomized_block_list, {}
-
-
 def _rewrite_read_stages(
     blocks: BlockList,
     stats: DatasetStats,
@@ -758,6 +748,8 @@ def _rewrite_read_stage(
         Non-lazy block list containing read tasks for not-yet-read block partitions,
         new stats for the block list, and the new list of stages.
     """
+    from ray.data._internal.stage_impl import RandomizeBlocksStage
+
     # Generate the "GetReadTasks" stage blocks.
     remote_args = in_blocks._remote_args
     blocks, metadata = [], []
@@ -798,6 +790,7 @@ def _reorder_stages(stages: List[Stage]) -> List[Stage]:
     Returns:
         Reordered stages.
     """
+    from ray.data._internal.stage_impl import RandomizeBlocksStage
 
     output: List[Stage] = []
     reorder_buf: List[RandomizeBlocksStage] = []

python/ray/data/_internal/stage_impl.py

@@ -0,0 +1,194 @@
+from typing import Optional, TYPE_CHECKING
+
+import ray
+from ray.data._internal.fast_repartition import fast_repartition
+from ray.data._internal.plan import AllToAllStage
+from ray.data._internal.shuffle_and_partition import (
+    PushBasedShufflePartitionOp,
+    SimpleShufflePartitionOp,
+)
+from ray.data._internal.block_list import BlockList
+from ray.data._internal.remote_fn import cached_remote_fn
+from ray.data._internal.sort import sort_impl
+from ray.data.context import DatasetContext
+from ray.data.block import (
+    _validate_key_fn,
+    Block,
+    KeyFn,
+    BlockMetadata,
+    BlockAccessor,
+    BlockExecStats,
+)
+
+if TYPE_CHECKING:
+    from ray.data import Dataset
+
+
+class RepartitionStage(AllToAllStage):
+    """Implementation of `Dataset.repartition()`."""
+
+    def __init__(self, num_blocks: int, shuffle: bool):
+        if shuffle:
+
+            def do_shuffle(
+                block_list, clear_input_blocks: bool, block_udf, remote_args
+            ):
+                if clear_input_blocks:
+                    blocks = block_list.copy()
+                    block_list.clear()
+                else:
+                    blocks = block_list
+                context = DatasetContext.get_current()
+                if context.use_push_based_shuffle:
+                    shuffle_op_cls = PushBasedShufflePartitionOp
+                else:
+                    shuffle_op_cls = SimpleShufflePartitionOp
+                shuffle_op = shuffle_op_cls(block_udf, random_shuffle=False)
+                return shuffle_op.execute(
+                    blocks,
+                    num_blocks,
+                    clear_input_blocks,
+                    map_ray_remote_args=remote_args,
+                    reduce_ray_remote_args=remote_args,
+                )
+
+            super().__init__(
+                "repartition", num_blocks, do_shuffle, supports_block_udf=True
+            )
+
+        else:
+
+            def do_fast_repartition(block_list, clear_input_blocks: bool, *_):
+                if clear_input_blocks:
+                    blocks = block_list.copy()
+                    block_list.clear()
+                else:
+                    blocks = block_list
+                return fast_repartition(blocks, num_blocks)
+
+            super().__init__("repartition", num_blocks, do_fast_repartition)
+
+
+class RandomizeBlocksStage(AllToAllStage):
+    """Implementation of `Dataset.randomize_blocks()`."""
+
+    def __init__(self, seed: Optional[int]):
+        self._seed = seed
+
+        super().__init__("randomize_block_order", None, self.do_randomize)
+
+    def do_randomize(self, block_list, *_):
+        num_blocks = block_list.initial_num_blocks()
+        if num_blocks == 0:
+            return block_list, {}
+        randomized_block_list = block_list.randomize_block_order(self._seed)
+        return randomized_block_list, {}
+
+
+class RandomShuffleStage(AllToAllStage):
+    """Implementation of `Dataset.random_shuffle()`."""
+
+    def __init__(self, seed: Optional[int], output_num_blocks: Optional[int]):
+        def do_shuffle(block_list, clear_input_blocks: bool, block_udf, remote_args):
+            num_blocks = block_list.executed_num_blocks()  # Blocking.
+            if num_blocks == 0:
+                return block_list, {}
+            if clear_input_blocks:
+                blocks = block_list.copy()
+                block_list.clear()
+            else:
+                blocks = block_list
+            context = DatasetContext.get_current()
+            if context.use_push_based_shuffle:
+                if output_num_blocks is not None:
+                    raise NotImplementedError(
+                        "Push-based shuffle doesn't support setting num_blocks yet."
+                    )
+                shuffle_op_cls = PushBasedShufflePartitionOp
+            else:
+                shuffle_op_cls = SimpleShufflePartitionOp
+            random_shuffle_op = shuffle_op_cls(
+                block_udf, random_shuffle=True, random_seed=seed
+            )
+            return random_shuffle_op.execute(
+                blocks,
+                output_num_blocks or num_blocks,
+                clear_input_blocks,
+                map_ray_remote_args=remote_args,
+                reduce_ray_remote_args=remote_args,
+            )
+
+        super().__init__(
+            "random_shuffle", output_num_blocks, do_shuffle, supports_block_udf=True
+        )
+
+
+class ZipStage(AllToAllStage):
+    """Implementation of `Dataset.zip()`."""
+
+    def __init__(self, other: "Dataset"):
+        def do_zip_all(block_list, clear_input_blocks: bool, *_):
+            blocks1 = block_list.get_blocks()
+            blocks2 = other.get_internal_block_refs()
+
+            if clear_input_blocks:
+                block_list.clear()
+
+            if len(blocks1) != len(blocks2):
+                # TODO(ekl) consider supporting if num_rows are equal.
+                raise ValueError(
+                    "Cannot zip dataset of different num blocks: {} vs {}".format(
+                        len(blocks1), len(blocks2)
+                    )
+                )
+
+            def do_zip(block1: Block, block2: Block) -> (Block, BlockMetadata):
+                stats = BlockExecStats.builder()
+                b1 = BlockAccessor.for_block(block1)
+                result = b1.zip(block2)
+                br = BlockAccessor.for_block(result)
+                return result, br.get_metadata(input_files=[], exec_stats=stats.build())
+
+            do_zip_fn = cached_remote_fn(do_zip, num_returns=2)
+
+            blocks = []
+            metadata = []
+            for b1, b2 in zip(blocks1, blocks2):
+                res, meta = do_zip_fn.remote(b1, b2)
+                blocks.append(res)
+                metadata.append(meta)
+
+            # Early release memory.
+            del blocks1, blocks2
+
+            # TODO(ekl) it might be nice to have a progress bar here.
+            metadata = ray.get(metadata)
+            blocks = BlockList(blocks, metadata)
+            return blocks, {}
+
+        super().__init__("zip", None, do_zip_all)
+
+
+class SortStage(AllToAllStage):
+    """Implementation of `Dataset.sort()`."""
+
+    def __init__(self, ds: "Dataset", key: Optional[KeyFn], descending: bool):
+        def do_sort(block_list, clear_input_blocks: bool, *_):
+            # Handle empty dataset.
+            if block_list.initial_num_blocks() == 0:
+                return block_list, {}
+            if clear_input_blocks:
+                blocks = block_list.copy()
+                block_list.clear()
+            else:
+                blocks = block_list
+            if isinstance(key, list):
+                if not key:
+                    raise ValueError("`key` must be a list of non-zero length")
+                for subkey in key:
+                    _validate_key_fn(ds, subkey)
+            else:
+                _validate_key_fn(ds, key)
+            return sort_impl(blocks, clear_input_blocks, key, descending)
+
+        super().__init__("sort", None, do_sort)

python/ray/data/dataset.py

@@ -32,22 +32,21 @@ from ray.data._internal.compute import (
     TaskPoolStrategy,
 )
 from ray.data._internal.delegating_block_builder import DelegatingBlockBuilder
-from ray.data._internal.fast_repartition import fast_repartition
 from ray.data._internal.lazy_block_list import LazyBlockList
 from ray.data._internal.output_buffer import BlockOutputBuffer
 from ray.data._internal.plan import (
-    AllToAllStage,
     ExecutionPlan,
     OneToOneStage,
+)
+from ray.data._internal.stage_impl import (
     RandomizeBlocksStage,
+    RepartitionStage,
+    RandomShuffleStage,
+    ZipStage,
+    SortStage,
 )
 from ray.data._internal.progress_bar import ProgressBar
 from ray.data._internal.remote_fn import cached_remote_fn
-from ray.data._internal.shuffle_and_partition import (
-    PushBasedShufflePartitionOp,
-    SimpleShufflePartitionOp,
-)
-from ray.data._internal.sort import sort_impl
 from ray.data._internal.stats import DatasetStats
 from ray.data._internal.table_block import VALUE_COL_NAME
 from ray.data.aggregate import AggregateFn, Max, Mean, Min, Std, Sum
@@ -723,50 +722,7 @@ class Dataset(Generic[T]):
             The repartitioned dataset.
         """
 
-        if shuffle:
-
-            def do_shuffle(
-                block_list, clear_input_blocks: bool, block_udf, remote_args
-            ):
-                if clear_input_blocks:
-                    blocks = block_list.copy()
-                    block_list.clear()
-                else:
-                    blocks = block_list
-                context = DatasetContext.get_current()
-                if context.use_push_based_shuffle:
-                    shuffle_op_cls = PushBasedShufflePartitionOp
-                else:
-                    shuffle_op_cls = SimpleShufflePartitionOp
-                shuffle_op = shuffle_op_cls(block_udf, random_shuffle=False)
-                return shuffle_op.execute(
-                    blocks,
-                    num_blocks,
-                    clear_input_blocks,
-                    map_ray_remote_args=remote_args,
-                    reduce_ray_remote_args=remote_args,
-                )
-
-            plan = self._plan.with_stage(
-                AllToAllStage(
-                    "repartition", num_blocks, do_shuffle, supports_block_udf=True
-                )
-            )
-
-        else:
-
-            def do_fast_repartition(block_list, clear_input_blocks: bool, *_):
-                if clear_input_blocks:
-                    blocks = block_list.copy()
-                    block_list.clear()
-                else:
-                    blocks = block_list
-                return fast_repartition(blocks, num_blocks)
-
-            plan = self._plan.with_stage(
-                AllToAllStage("repartition", num_blocks, do_fast_repartition)
-            )
-
+        plan = self._plan.with_stage(RepartitionStage(num_blocks, shuffle))
         return Dataset(plan, self._epoch, self._lazy)
 
     def random_shuffle(
@@ -799,36 +755,7 @@ class Dataset(Generic[T]):
             The shuffled dataset.
         """
 
-        def do_shuffle(block_list, clear_input_blocks: bool, block_udf, remote_args):
-            num_blocks = block_list.executed_num_blocks()  # Blocking.
-            if num_blocks == 0:
-                return block_list, {}
-            if clear_input_blocks:
-                blocks = block_list.copy()
-                block_list.clear()
-            else:
-                blocks = block_list
-            context = DatasetContext.get_current()
-            if context.use_push_based_shuffle:
-                shuffle_op_cls = PushBasedShufflePartitionOp
-            else:
-                shuffle_op_cls = SimpleShufflePartitionOp
-            random_shuffle_op = shuffle_op_cls(
-                block_udf, random_shuffle=True, random_seed=seed
-            )
-            return random_shuffle_op.execute(
-                blocks,
-                num_blocks,
-                clear_input_blocks,
-                map_ray_remote_args=remote_args,
-                reduce_ray_remote_args=remote_args,
-            )
-
-        plan = self._plan.with_stage(
-            AllToAllStage(
-                "random_shuffle", num_blocks, do_shuffle, supports_block_udf=True
-            )
-        )
+        plan = self._plan.with_stage(RandomShuffleStage(seed, num_blocks))
         return Dataset(plan, self._epoch, self._lazy)
 
     def randomize_block_order(
@@ -1834,25 +1761,7 @@ class Dataset(Generic[T]):
             A new, sorted dataset.
         """
 
-        def do_sort(block_list, clear_input_blocks: bool, *_):
-            # Handle empty dataset.
-            if block_list.initial_num_blocks() == 0:
-                return block_list, {}
-            if clear_input_blocks:
-                blocks = block_list.copy()
-                block_list.clear()
-            else:
-                blocks = block_list
-            if isinstance(key, list):
-                if not key:
-                    raise ValueError("`key` must be a list of non-zero length")
-                for subkey in key:
-                    _validate_key_fn(self, subkey)
-            else:
-                _validate_key_fn(self, key)
-            return sort_impl(blocks, clear_input_blocks, key, descending)
-
-        plan = self._plan.with_stage(AllToAllStage("sort", None, do_sort))
+        plan = self._plan.with_stage(SortStage(self, key, descending))
         return Dataset(plan, self._epoch, self._lazy)
 
     def zip(self, other: "Dataset[U]") -> "Dataset[(T, U)]":
@@ -1882,46 +1791,7 @@ class Dataset(Generic[T]):
             comes from the first dataset and v comes from the second.
         """
 
-        def do_zip_all(block_list, clear_input_blocks: bool, *_):
-            blocks1 = block_list.get_blocks()
-            blocks2 = other.get_internal_block_refs()
-
-            if clear_input_blocks:
-                block_list.clear()
-
-            if len(blocks1) != len(blocks2):
-                # TODO(ekl) consider supporting if num_rows are equal.
-                raise ValueError(
-                    "Cannot zip dataset of different num blocks: {} vs {}".format(
-                        len(blocks1), len(blocks2)
-                    )
-                )
-
-            def do_zip(block1: Block, block2: Block) -> (Block, BlockMetadata):
-                stats = BlockExecStats.builder()
-                b1 = BlockAccessor.for_block(block1)
-                result = b1.zip(block2)
-                br = BlockAccessor.for_block(result)
-                return result, br.get_metadata(input_files=[], exec_stats=stats.build())
-
-            do_zip_fn = cached_remote_fn(do_zip, num_returns=2)
-
-            blocks = []
-            metadata = []
-            for b1, b2 in zip(blocks1, blocks2):
-                res, meta = do_zip_fn.remote(b1, b2)
-                blocks.append(res)
-                metadata.append(meta)
-
-            # Early release memory.
-            del blocks1, blocks2
-
-            # TODO(ekl) it might be nice to have a progress bar here.
-            metadata = ray.get(metadata)
-            blocks = BlockList(blocks, metadata)
-            return blocks, {}
-
-        plan = self._plan.with_stage(AllToAllStage("zip", None, do_zip_all))
+        plan = self._plan.with_stage(ZipStage(other))
         return Dataset(plan, self._epoch, self._lazy)
 
     def limit(self, limit: int) -> "Dataset[T]":

python/ray/data/tests/test_dataset.py

@@ -3975,9 +3975,13 @@ def test_random_shuffle(shutdown_only, pipelined, use_push_based_shuffle):
     r2 = range(100, parallelism=1).random_shuffle().take(999)
     assert r1 != r2, (r1, r2)
 
-    r1 = range(100).random_shuffle(num_blocks=1).take(999)
-    r2 = range(100).random_shuffle(num_blocks=1).take(999)
-    assert r1 != r2, (r1, r2)
+    # TODO(swang): fix this
+    if not use_push_based_shuffle:
+        if not pipelined:
+            assert range(100).random_shuffle(num_blocks=1).num_blocks() == 1
+        r1 = range(100).random_shuffle(num_blocks=1).take(999)
+        r2 = range(100).random_shuffle(num_blocks=1).take(999)
+        assert r1 != r2, (r1, r2)
 
     r0 = range(100, parallelism=5).take(999)
     r1 = range(100, parallelism=5).random_shuffle(seed=0).take(999)