xintongsong commented on code in PR #22501: URL: https://github.com/apache/flink/pull/22501#discussion_r1208271941
########## flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/hybrid/tiered/storage/SubpartitionHashBufferAccumulator.java: ########## @@ -0,0 +1,169 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.flink.runtime.io.network.partition.hybrid.tiered.storage; + +import org.apache.flink.core.memory.MemorySegment; +import org.apache.flink.core.memory.MemorySegmentFactory; +import org.apache.flink.runtime.io.network.buffer.Buffer; +import org.apache.flink.runtime.io.network.buffer.BufferBuilder; +import org.apache.flink.runtime.io.network.buffer.BufferConsumer; +import org.apache.flink.runtime.io.network.buffer.FreeingBufferRecycler; +import org.apache.flink.runtime.io.network.buffer.NetworkBuffer; +import org.apache.flink.runtime.io.network.partition.hybrid.tiered.common.TieredStorageSubpartitionId; + +import java.io.IOException; +import java.nio.ByteBuffer; +import java.util.Collections; +import java.util.LinkedList; +import java.util.List; +import java.util.Optional; +import java.util.Queue; +import java.util.function.BiConsumer; + +import static org.apache.flink.util.Preconditions.checkArgument; +import static org.apache.flink.util.Preconditions.checkNotNull; +import static org.apache.flink.util.Preconditions.checkState; + +/** + * {@link SubpartitionHashBufferAccumulator} accumulates the records in a subpartition. + * + * <p>Note that {@link #setup} need an argument of buffer flush listener to accept the finished + * accumulated buffers. + */ +public class SubpartitionHashBufferAccumulator { + + private final TieredStorageSubpartitionId subpartitionId; + + private final int bufferSize; + + private final HashBufferAccumulatorOperation bufferAccumulatorOperation; + + private BiConsumer<TieredStorageSubpartitionId, List<Buffer>> accumulatedBufferFlusher; Review Comment: Same here for the nullable field. But this can be simplified if we remove `setup`. ########## flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/hybrid/tiered/storage/HashBufferAccumulator.java: ########## @@ -0,0 +1,95 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.flink.runtime.io.network.partition.hybrid.tiered.storage; + +import org.apache.flink.runtime.io.network.buffer.Buffer; +import org.apache.flink.runtime.io.network.buffer.BufferBuilder; +import org.apache.flink.runtime.io.network.partition.hybrid.tiered.common.TieredStorageSubpartitionId; +import org.apache.flink.util.ExceptionUtils; + +import java.io.IOException; +import java.nio.ByteBuffer; +import java.util.Arrays; +import java.util.List; +import java.util.function.BiConsumer; + +/** + * The hash implementation of the {@link BufferAccumulator}. The {@link BufferAccumulator} receives + * the records from {@link TieredStorageProducerClient} and the records will accumulate and + * transform to finished buffers. The finished buffers will be transferred to the corresponding tier + * dynamically. + */ +public class HashBufferAccumulator implements BufferAccumulator, HashBufferAccumulatorOperation { + + private final int bufferSize; + + private final TieredStorageMemoryManager storageMemoryManager; + + private SubpartitionHashBufferAccumulator[] subpartitionHashBufferAccumulators; Review Comment: This can be null before setting up. We should: - Annotate it with `@Nullable` - Explain the semantics for null-value - Check before using it ########## flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/hybrid/tiered/storage/HashBufferAccumulator.java: ########## @@ -0,0 +1,95 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.flink.runtime.io.network.partition.hybrid.tiered.storage; + +import org.apache.flink.runtime.io.network.buffer.Buffer; +import org.apache.flink.runtime.io.network.buffer.BufferBuilder; +import org.apache.flink.runtime.io.network.partition.hybrid.tiered.common.TieredStorageSubpartitionId; +import org.apache.flink.util.ExceptionUtils; + +import java.io.IOException; +import java.nio.ByteBuffer; +import java.util.Arrays; +import java.util.List; +import java.util.function.BiConsumer; + +/** + * The hash implementation of the {@link BufferAccumulator}. The {@link BufferAccumulator} receives + * the records from {@link TieredStorageProducerClient} and the records will accumulate and + * transform to finished buffers. The finished buffers will be transferred to the corresponding tier + * dynamically. + */ Review Comment: For the accumulators, I think it's important to explain how many memory it requires and when are the buffers flushed, because these are the major differences between hash- and sort-based accumulators. ########## flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/hybrid/tiered/storage/SubpartitionHashBufferAccumulator.java: ########## @@ -0,0 +1,169 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.flink.runtime.io.network.partition.hybrid.tiered.storage; + +import org.apache.flink.core.memory.MemorySegment; +import org.apache.flink.core.memory.MemorySegmentFactory; +import org.apache.flink.runtime.io.network.buffer.Buffer; +import org.apache.flink.runtime.io.network.buffer.BufferBuilder; +import org.apache.flink.runtime.io.network.buffer.BufferConsumer; +import org.apache.flink.runtime.io.network.buffer.FreeingBufferRecycler; +import org.apache.flink.runtime.io.network.buffer.NetworkBuffer; +import org.apache.flink.runtime.io.network.partition.hybrid.tiered.common.TieredStorageSubpartitionId; + +import java.io.IOException; +import java.nio.ByteBuffer; +import java.util.Collections; +import java.util.LinkedList; +import java.util.List; +import java.util.Optional; +import java.util.Queue; +import java.util.function.BiConsumer; + +import static org.apache.flink.util.Preconditions.checkArgument; +import static org.apache.flink.util.Preconditions.checkNotNull; +import static org.apache.flink.util.Preconditions.checkState; + +/** + * {@link SubpartitionHashBufferAccumulator} accumulates the records in a subpartition. + * + * <p>Note that {@link #setup} need an argument of buffer flush listener to accept the finished + * accumulated buffers. + */ +public class SubpartitionHashBufferAccumulator { + + private final TieredStorageSubpartitionId subpartitionId; + + private final int bufferSize; + + private final HashBufferAccumulatorOperation bufferAccumulatorOperation; + + private BiConsumer<TieredStorageSubpartitionId, List<Buffer>> accumulatedBufferFlusher; + + // Not guarded by lock because it is expected only accessed from task's main thread. Review Comment: Instead of commenting here, we may claim the class is not thread-safe in java doc. IIUC, the entire class, in addition to this queue, should be accessed only from the main thread? ########## flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/hybrid/tiered/storage/SubpartitionHashBufferAccumulator.java: ########## @@ -0,0 +1,169 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.flink.runtime.io.network.partition.hybrid.tiered.storage; + +import org.apache.flink.core.memory.MemorySegment; +import org.apache.flink.core.memory.MemorySegmentFactory; +import org.apache.flink.runtime.io.network.buffer.Buffer; +import org.apache.flink.runtime.io.network.buffer.BufferBuilder; +import org.apache.flink.runtime.io.network.buffer.BufferConsumer; +import org.apache.flink.runtime.io.network.buffer.FreeingBufferRecycler; +import org.apache.flink.runtime.io.network.buffer.NetworkBuffer; +import org.apache.flink.runtime.io.network.partition.hybrid.tiered.common.TieredStorageSubpartitionId; + +import java.io.IOException; +import java.nio.ByteBuffer; +import java.util.Collections; +import java.util.LinkedList; +import java.util.List; +import java.util.Optional; +import java.util.Queue; +import java.util.function.BiConsumer; + +import static org.apache.flink.util.Preconditions.checkArgument; +import static org.apache.flink.util.Preconditions.checkNotNull; +import static org.apache.flink.util.Preconditions.checkState; + +/** + * {@link SubpartitionHashBufferAccumulator} accumulates the records in a subpartition. + * + * <p>Note that {@link #setup} need an argument of buffer flush listener to accept the finished + * accumulated buffers. + */ +public class SubpartitionHashBufferAccumulator { Review Comment: I'd suggest the name `HashSubpartitionBufferAccumulator`. ########## flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/hybrid/tiered/storage/HashBufferAccumulatorOperation.java: ########## @@ -0,0 +1,36 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.flink.runtime.io.network.partition.hybrid.tiered.storage; + +import org.apache.flink.runtime.io.network.buffer.BufferBuilder; +import org.apache.flink.runtime.io.network.buffer.BufferPool; + +/** + * This interface is used by {@link SubpartitionHashBufferAccumulator} to operate {@link + * HashBufferAccumulator}. + */ +public interface HashBufferAccumulatorOperation { Review Comment: I'd suggest the name `HashSubpartitionBufferAccumulatorContext`. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
