ilooner commented on a change in pull request #1606: DRILL-6845: Semi-Hash-Join to skip incoming build duplicates, automatically stop skipping if too few URL: https://github.com/apache/drill/pull/1606#discussion_r248152865
########## File path: exec/java-exec/src/main/java/org/apache/drill/exec/physical/impl/join/HashJoinSpillControlImpl.java ########## @@ -0,0 +1,238 @@ +/* + * 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.drill.exec.physical.impl.join; + + import org.apache.drill.exec.ExecConstants; + import org.apache.drill.exec.memory.BufferAllocator; + import org.apache.drill.exec.ops.FragmentContext; + import org.apache.drill.exec.record.RecordBatch; + import org.apache.drill.exec.record.RecordBatchMemoryManager; + import org.slf4j.Logger; + import org.slf4j.LoggerFactory; + + import javax.annotation.Nullable; + import java.util.Set; + + import static org.apache.drill.exec.record.JoinBatchMemoryManager.LEFT_INDEX; + import static org.apache.drill.exec.record.JoinBatchMemoryManager.RIGHT_INDEX; + +/** + * This class is currently used only for Semi-Hash-Join that avoids duplicates by the use of a hash table + * The method {@link HashJoinMemoryCalculator.HashJoinSpillControl#shouldSpill()} returns true if the memory available now to the allocator if not enough + * to hold (a multiple of, for safety) a new allocated batch + */ +public class HashJoinSpillControlImpl implements HashJoinMemoryCalculator.BuildSidePartitioning { + private static final Logger logger = LoggerFactory.getLogger(HashJoinSpillControlImpl.class); + + private BufferAllocator allocator; + private int recordsPerBatch; + private int minBatchesInAvailableMemory; + private RecordBatchMemoryManager batchMemoryManager; + private int initialPartitions; + private int numPartitions; + private int recordsPerPartitionBatchProbe; + private FragmentContext context; + HashJoinMemoryCalculator.PartitionStatSet partitionStatSet; + + HashJoinSpillControlImpl(BufferAllocator allocator, int recordsPerBatch, int minBatchesInAvailableMemory, RecordBatchMemoryManager batchMemoryManager, FragmentContext context) { + this.allocator = allocator; + this.recordsPerBatch = recordsPerBatch; + this.minBatchesInAvailableMemory = minBatchesInAvailableMemory; + this.batchMemoryManager = batchMemoryManager; + this.context = context; + } + + @Override + public boolean shouldSpill() { + // Expected new batch size like the current, plus the Hash Values vector (4 bytes per HV) + long batchSize = ( batchMemoryManager.getRecordBatchSizer(RIGHT_INDEX).getRowAllocWidth() + 4 ) * recordsPerBatch; + long reserveForOutgoing = batchMemoryManager.getOutputBatchSize(); + long memoryAvailableNow = allocator.getLimit() - allocator.getAllocatedMemory() - reserveForOutgoing; + boolean needsSpill = minBatchesInAvailableMemory * batchSize > memoryAvailableNow; + if ( needsSpill ) { + logger.debug("should spill now - batch size {}, mem avail {}, reserved for outgoing {}", batchSize, memoryAvailableNow, reserveForOutgoing); + } + return needsSpill; // go spill if too little memory is available + } + + @Override + public void initialize(boolean firstCycle, + boolean reserveHash, + RecordBatch buildSideBatch, + RecordBatch probeSideBatch, + Set<String> joinColumns, + boolean probeEmpty, + long memoryAvailable, + int initialPartitions, + int recordsPerPartitionBatchBuild, + int recordsPerPartitionBatchProbe, + int maxBatchNumRecordsBuild, + int maxBatchNumRecordsProbe, + int outputBatchSize, + double loadFactor) { + this.initialPartitions = initialPartitions; + this.recordsPerPartitionBatchProbe = recordsPerPartitionBatchProbe; + + calculateMemoryUsage(); + } + + @Override + public void setPartitionStatSet(HashJoinMemoryCalculator.PartitionStatSet partitionStatSet) { + this.partitionStatSet = partitionStatSet; + } + + @Override + public int getNumPartitions() { + return numPartitions; + } + + /** + * Calculate the number of partitions possible for the given available memory + * start at initialPartitions and adjust down (in powers of 2) as needed + */ + private void calculateMemoryUsage() { + long reserveForOutgoing = batchMemoryManager.getOutputBatchSize(); + long memoryAvailableNow = allocator.getLimit() - allocator.getAllocatedMemory() - reserveForOutgoing; + + // Expected new batch size like the current, plus the Hash Values vector (4 bytes per HV) + int buildBatchSize = ( batchMemoryManager.getRecordBatchSizer(RIGHT_INDEX).getRowAllocWidth() + 4 ) * recordsPerBatch; + int hashTableSize = buildBatchSize /* the keys in the HT */ + Review comment: Please reuse the HashTableSizeCalculator here. It has already been unit tested and has the benefit of using the actual key sizes. So the user will not be required to provide the `ExecConstants.MIN_HASH_TABLE_SIZE_KEY`. For an example of how to use it see **HashJoinMemoryCalculatorImpl line 766**. And for an example of how to compute the key sizes see **HashJoinMemoryCalculatorImpl line 264 - 268**. ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services
