[HUDI-2101][RFC-28] support z-order for hudi (#3330)
* [HUDI-2101]support z-order for hudi * Renaming some configs for consistency/simplicity. * Minor code cleanups Co-authored-by: Vinoth Chandar <vinoth@apache.org>
This commit is contained in:
xiarixiaoyao
@@ -380,6 +380,10 @@ public class SparkRDDWriteClient<T extends HoodieRecordPayload> extends
|
||||
writeTableMetadata(table, metadata, new HoodieInstant(HoodieInstant.State.INFLIGHT, HoodieTimeline.REPLACE_COMMIT_ACTION, clusteringCommitTime));
|
||||
finalizeWrite(table, clusteringCommitTime, writeStats);
|
||||
try {
|
||||
// try to save statistics info to hudi
|
||||
if (config.isDataSkippingEnabled() && config.isLayoutOptimizationEnabled() && !config.getClusteringSortColumns().isEmpty()) {
|
||||
table.updateStatistics(context, writeStats, clusteringCommitTime, true);
|
||||
}
|
||||
LOG.info("Committing Clustering " + clusteringCommitTime + ". Finished with result " + metadata);
|
||||
table.getActiveTimeline().transitionReplaceInflightToComplete(
|
||||
HoodieTimeline.getReplaceCommitInflightInstant(clusteringCommitTime),
|
||||
|
||||
@@ -20,6 +20,7 @@ package org.apache.hudi.client.clustering.run.strategy;
|
||||
|
||||
import org.apache.hudi.avro.HoodieAvroUtils;
|
||||
import org.apache.hudi.client.WriteStatus;
|
||||
import org.apache.hudi.client.common.HoodieSparkEngineContext;
|
||||
import org.apache.hudi.common.engine.HoodieEngineContext;
|
||||
import org.apache.hudi.common.model.HoodieFileGroupId;
|
||||
import org.apache.hudi.common.model.HoodieRecord;
|
||||
@@ -28,6 +29,7 @@ import org.apache.hudi.common.util.Option;
|
||||
import org.apache.hudi.config.HoodieStorageConfig;
|
||||
import org.apache.hudi.config.HoodieWriteConfig;
|
||||
import org.apache.hudi.execution.bulkinsert.RDDCustomColumnsSortPartitioner;
|
||||
import org.apache.hudi.execution.bulkinsert.RDDSpatialCurveOptimizationSortPartitioner;
|
||||
import org.apache.hudi.table.BulkInsertPartitioner;
|
||||
import org.apache.hudi.table.HoodieTable;
|
||||
import org.apache.hudi.table.action.commit.SparkBulkInsertHelper;
|
||||
@@ -77,7 +79,11 @@ public class SparkSortAndSizeExecutionStrategy<T extends HoodieRecordPayload<T>>
|
||||
* Create BulkInsertPartitioner based on strategy params.
|
||||
*/
|
||||
protected Option<BulkInsertPartitioner<T>> getPartitioner(Map<String, String> strategyParams, Schema schema) {
|
||||
if (strategyParams.containsKey(PLAN_STRATEGY_SORT_COLUMNS.key())) {
|
||||
if (getWriteConfig().isLayoutOptimizationEnabled()) {
|
||||
// sort input records by z-order/hilbert
|
||||
return Option.of(new RDDSpatialCurveOptimizationSortPartitioner((HoodieSparkEngineContext) getEngineContext(),
|
||||
getWriteConfig(), HoodieAvroUtils.addMetadataFields(schema)));
|
||||
} else if (strategyParams.containsKey(PLAN_STRATEGY_SORT_COLUMNS.key())) {
|
||||
return Option.of(new RDDCustomColumnsSortPartitioner(strategyParams.get(PLAN_STRATEGY_SORT_COLUMNS.key()).split(","),
|
||||
HoodieAvroUtils.addMetadataFields(schema)));
|
||||
} else {
|
||||
|
||||
@@ -0,0 +1,98 @@
|
||||
/*
|
||||
* 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.hudi.execution.bulkinsert;
|
||||
|
||||
import org.apache.hudi.AvroConversionUtils;
|
||||
import org.apache.hudi.HoodieSparkUtils;
|
||||
import org.apache.hudi.client.common.HoodieSparkEngineContext;
|
||||
import org.apache.hudi.common.config.SerializableSchema;
|
||||
import org.apache.hudi.common.model.HoodieKey;
|
||||
import org.apache.hudi.common.model.HoodieRecord;
|
||||
import org.apache.hudi.common.model.HoodieRecordPayload;
|
||||
import org.apache.hudi.common.util.Option;
|
||||
import org.apache.hudi.common.util.ReflectionUtils;
|
||||
import org.apache.hudi.config.HoodieWriteConfig;
|
||||
import org.apache.hudi.exception.HoodieException;
|
||||
import org.apache.hudi.table.BulkInsertPartitioner;
|
||||
|
||||
import org.apache.avro.Schema;
|
||||
import org.apache.avro.generic.GenericRecord;
|
||||
import org.apache.spark.ZCurveOptimizeHelper;
|
||||
import org.apache.spark.api.java.JavaRDD;
|
||||
import org.apache.spark.sql.Dataset;
|
||||
import org.apache.spark.sql.Row;
|
||||
|
||||
/**
|
||||
* A partitioner that does spartial curve optimization sorting based on specified column values for each RDD partition.
|
||||
* support z-curve optimization, hilbert will come soon.
|
||||
* @param <T> HoodieRecordPayload type
|
||||
*/
|
||||
public class RDDSpatialCurveOptimizationSortPartitioner<T extends HoodieRecordPayload>
|
||||
implements BulkInsertPartitioner<JavaRDD<HoodieRecord<T>>> {
|
||||
private final HoodieSparkEngineContext sparkEngineContext;
|
||||
private final SerializableSchema serializableSchema;
|
||||
private final HoodieWriteConfig config;
|
||||
|
||||
public RDDSpatialCurveOptimizationSortPartitioner(HoodieSparkEngineContext sparkEngineContext, HoodieWriteConfig config, Schema schema) {
|
||||
this.sparkEngineContext = sparkEngineContext;
|
||||
this.config = config;
|
||||
this.serializableSchema = new SerializableSchema(schema);
|
||||
}
|
||||
|
||||
@Override
|
||||
public JavaRDD<HoodieRecord<T>> repartitionRecords(JavaRDD<HoodieRecord<T>> records, int outputSparkPartitions) {
|
||||
String payloadClass = config.getPayloadClass();
|
||||
// do sort
|
||||
JavaRDD<GenericRecord> preparedRecord = prepareGenericRecord(records, outputSparkPartitions, serializableSchema.get());
|
||||
return preparedRecord.map(record -> {
|
||||
String key = record.get(HoodieRecord.RECORD_KEY_METADATA_FIELD).toString();
|
||||
String partition = record.get(HoodieRecord.PARTITION_PATH_METADATA_FIELD).toString();
|
||||
HoodieKey hoodieKey = new HoodieKey(key, partition);
|
||||
HoodieRecordPayload avroPayload = ReflectionUtils.loadPayload(payloadClass,
|
||||
new Object[] {Option.of(record)}, Option.class);
|
||||
HoodieRecord hoodieRecord = new HoodieRecord(hoodieKey, avroPayload);
|
||||
return hoodieRecord;
|
||||
});
|
||||
}
|
||||
|
||||
private JavaRDD<GenericRecord> prepareGenericRecord(JavaRDD<HoodieRecord<T>> inputRecords, final int numOutputGroups, final Schema schema) {
|
||||
SerializableSchema serializableSchema = new SerializableSchema(schema);
|
||||
JavaRDD<GenericRecord> genericRecordJavaRDD = inputRecords.map(f -> (GenericRecord) f.getData().getInsertValue(serializableSchema.get()).get());
|
||||
Dataset<Row> originDF = AvroConversionUtils.createDataFrame(genericRecordJavaRDD.rdd(), schema.toString(), sparkEngineContext.getSqlContext().sparkSession());
|
||||
Dataset<Row> zDataFrame;
|
||||
|
||||
switch (config.getLayoutOptimizationCurveBuildMethod()) {
|
||||
case DIRECT:
|
||||
zDataFrame = ZCurveOptimizeHelper.createZIndexedDataFrameByMapValue(originDF, config.getClusteringSortColumns(), numOutputGroups);
|
||||
break;
|
||||
case SAMPLE:
|
||||
zDataFrame = ZCurveOptimizeHelper.createZIndexedDataFrameBySample(originDF, config.getClusteringSortColumns(), numOutputGroups);
|
||||
break;
|
||||
default:
|
||||
throw new HoodieException("Not a valid build curve method for doWriteOperation: ");
|
||||
}
|
||||
return HoodieSparkUtils.createRdd(zDataFrame, schema.getName(),
|
||||
schema.getNamespace(), false, org.apache.hudi.common.util.Option.empty()).toJavaRDD();
|
||||
}
|
||||
|
||||
@Override
|
||||
public boolean arePartitionRecordsSorted() {
|
||||
return true;
|
||||
}
|
||||
}
|
||||
@@ -18,6 +18,7 @@
|
||||
|
||||
package org.apache.hudi.table;
|
||||
|
||||
import org.apache.hadoop.fs.Path;
|
||||
import org.apache.hudi.avro.model.HoodieCleanMetadata;
|
||||
import org.apache.hudi.avro.model.HoodieCleanerPlan;
|
||||
import org.apache.hudi.avro.model.HoodieClusteringPlan;
|
||||
@@ -34,6 +35,7 @@ import org.apache.hudi.common.model.HoodieBaseFile;
|
||||
import org.apache.hudi.common.model.HoodieKey;
|
||||
import org.apache.hudi.common.model.HoodieRecord;
|
||||
import org.apache.hudi.common.model.HoodieRecordPayload;
|
||||
import org.apache.hudi.common.model.HoodieWriteStat;
|
||||
import org.apache.hudi.common.table.HoodieTableMetaClient;
|
||||
import org.apache.hudi.common.table.timeline.HoodieInstant;
|
||||
import org.apache.hudi.common.table.timeline.HoodieTimeline;
|
||||
@@ -71,13 +73,16 @@ import org.apache.hudi.table.action.rollback.CopyOnWriteRollbackActionExecutor;
|
||||
import org.apache.hudi.table.action.savepoint.SavepointActionExecutor;
|
||||
import org.apache.log4j.LogManager;
|
||||
import org.apache.log4j.Logger;
|
||||
import org.apache.spark.ZCurveOptimizeHelper;
|
||||
import org.apache.spark.api.java.JavaRDD;
|
||||
import scala.collection.JavaConversions;
|
||||
|
||||
import java.io.IOException;
|
||||
import java.util.Collections;
|
||||
import java.util.Iterator;
|
||||
import java.util.List;
|
||||
import java.util.Map;
|
||||
import java.util.stream.Collectors;
|
||||
|
||||
/**
|
||||
* Implementation of a very heavily read-optimized Hoodie Table where, all data is stored in base files, with
|
||||
@@ -152,6 +157,32 @@ public class HoodieSparkCopyOnWriteTable<T extends HoodieRecordPayload>
|
||||
return new SparkInsertOverwriteTableCommitActionExecutor(context, config, this, instantTime, records).execute();
|
||||
}
|
||||
|
||||
@Override
|
||||
public void updateStatistics(HoodieEngineContext context, List<HoodieWriteStat> stats, String instantTime, Boolean isOptimizeOperation) {
|
||||
// deal with z-order/hilbert statistic info
|
||||
if (isOptimizeOperation) {
|
||||
updateOptimizeOperationStatistics(context, stats, instantTime);
|
||||
}
|
||||
}
|
||||
|
||||
private void updateOptimizeOperationStatistics(HoodieEngineContext context, List<HoodieWriteStat> stats, String instantTime) {
|
||||
String cols = config.getClusteringSortColumns();
|
||||
String basePath = metaClient.getBasePath();
|
||||
String indexPath = metaClient.getZindexPath();
|
||||
List<String> validateCommits = metaClient.getCommitsTimeline()
|
||||
.filterCompletedInstants().getInstants().map(f -> f.getTimestamp()).collect(Collectors.toList());
|
||||
List<String> touchFiles = stats.stream().map(s -> new Path(basePath, s.getPath()).toString()).collect(Collectors.toList());
|
||||
if (touchFiles.isEmpty() || cols.isEmpty() || indexPath.isEmpty()) {
|
||||
LOG.warn("save nothing to index table");
|
||||
return;
|
||||
}
|
||||
HoodieSparkEngineContext sparkEngineContext = (HoodieSparkEngineContext)context;
|
||||
ZCurveOptimizeHelper.saveStatisticsInfo(sparkEngineContext
|
||||
.getSqlContext().sparkSession().read().load(JavaConversions.asScalaBuffer(touchFiles)),
|
||||
cols, indexPath, instantTime, validateCommits);
|
||||
LOG.info(String.format("save statistic info sucessfully at commitTime: %s", instantTime));
|
||||
}
|
||||
|
||||
@Override
|
||||
public Option<HoodieCompactionPlan> scheduleCompaction(HoodieEngineContext context, String instantTime, Option<Map<String, String>> extraMetadata) {
|
||||
throw new HoodieNotSupportedException("Compaction is not supported on a CopyOnWrite table");
|
||||
|
||||
@@ -69,7 +69,7 @@ public class SparkBulkInsertHelper<T extends HoodieRecordPayload, R> extends Abs
|
||||
|
||||
//transition bulk_insert state to inflight
|
||||
table.getActiveTimeline().transitionRequestedToInflight(new HoodieInstant(HoodieInstant.State.REQUESTED,
|
||||
table.getMetaClient().getCommitActionType(), instantTime), Option.empty(),
|
||||
executor.getCommitActionType(), instantTime), Option.empty(),
|
||||
config.shouldAllowMultiWriteOnSameInstant());
|
||||
// write new files
|
||||
JavaRDD<WriteStatus> writeStatuses = bulkInsert(inputRecords, instantTime, table, config, performDedupe, userDefinedBulkInsertPartitioner, false, config.getBulkInsertShuffleParallelism(), false);
|
||||
|
||||
@@ -0,0 +1,355 @@
|
||||
/*
|
||||
* 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.spark;
|
||||
|
||||
import scala.collection.JavaConversions;
|
||||
|
||||
import org.apache.hadoop.conf.Configuration;
|
||||
import org.apache.hadoop.fs.FileSystem;
|
||||
import org.apache.hadoop.fs.Path;
|
||||
import org.apache.hudi.HoodieSparkUtils$;
|
||||
import org.apache.hudi.common.fs.FSUtils;
|
||||
import org.apache.hudi.common.model.HoodieColumnRangeMetadata;
|
||||
import org.apache.hudi.common.model.HoodieFileFormat;
|
||||
import org.apache.hudi.common.util.BaseFileUtils;
|
||||
import org.apache.hudi.common.util.Option;
|
||||
import org.apache.hudi.common.util.ParquetUtils;
|
||||
import org.apache.hudi.exception.HoodieException;
|
||||
import org.apache.hudi.optimize.ZOrderingUtil;
|
||||
import org.apache.parquet.io.api.Binary;
|
||||
import org.apache.spark.api.java.JavaRDD;
|
||||
import org.apache.spark.api.java.JavaSparkContext;
|
||||
import org.apache.spark.sql.Dataset;
|
||||
import org.apache.spark.sql.Row;
|
||||
import org.apache.spark.sql.Row$;
|
||||
import org.apache.spark.sql.SparkSession;
|
||||
import org.apache.spark.sql.hudi.execution.RangeSampleSort$;
|
||||
import org.apache.spark.sql.hudi.execution.ZorderingBinarySort;
|
||||
import org.apache.spark.sql.types.BinaryType;
|
||||
import org.apache.spark.sql.types.BinaryType$;
|
||||
import org.apache.spark.sql.types.BooleanType;
|
||||
import org.apache.spark.sql.types.ByteType;
|
||||
import org.apache.spark.sql.types.DataType;
|
||||
import org.apache.spark.sql.types.DateType;
|
||||
import org.apache.spark.sql.types.DecimalType;
|
||||
import org.apache.spark.sql.types.DoubleType;
|
||||
import org.apache.spark.sql.types.FloatType;
|
||||
import org.apache.spark.sql.types.IntegerType;
|
||||
import org.apache.spark.sql.types.LongType;
|
||||
import org.apache.spark.sql.types.LongType$;
|
||||
import org.apache.spark.sql.types.Metadata;
|
||||
import org.apache.spark.sql.types.ShortType;
|
||||
import org.apache.spark.sql.types.StringType;
|
||||
import org.apache.spark.sql.types.StringType$;
|
||||
import org.apache.spark.sql.types.StructField;
|
||||
import org.apache.spark.sql.types.StructType$;
|
||||
import org.apache.spark.sql.types.TimestampType;
|
||||
import org.apache.spark.util.SerializableConfiguration;
|
||||
|
||||
import java.io.IOException;
|
||||
import java.math.BigDecimal;
|
||||
import java.util.ArrayList;
|
||||
import java.util.Arrays;
|
||||
import java.util.Collection;
|
||||
import java.util.List;
|
||||
import java.util.Map;
|
||||
import java.util.stream.Collectors;
|
||||
|
||||
public class ZCurveOptimizeHelper {
|
||||
|
||||
private static final String SPARK_JOB_DESCRIPTION = "spark.job.description";
|
||||
|
||||
/**
|
||||
* Create z-order DataFrame directly
|
||||
* first, map all base type data to byte[8], then create z-order DataFrame
|
||||
* only support base type data. long,int,short,double,float,string,timestamp,decimal,date,byte
|
||||
* this method is more effective than createZIndexDataFrameBySample
|
||||
*
|
||||
* @param df a spark DataFrame holds parquet files to be read.
|
||||
* @param zCols z-sort cols
|
||||
* @param fileNum spark partition num
|
||||
* @return a dataFrame sorted by z-order.
|
||||
*/
|
||||
public static Dataset<Row> createZIndexedDataFrameByMapValue(Dataset<Row> df, List<String> zCols, int fileNum) {
|
||||
Map<String, StructField> columnsMap = Arrays.stream(df.schema().fields()).collect(Collectors.toMap(e -> e.name(), e -> e));
|
||||
int fieldNum = df.schema().fields().length;
|
||||
List<String> checkCols = zCols.stream().filter(f -> columnsMap.containsKey(f)).collect(Collectors.toList());
|
||||
if (zCols.size() != checkCols.size()) {
|
||||
return df;
|
||||
}
|
||||
// only one col to sort, no need to use z-order
|
||||
if (zCols.size() == 1) {
|
||||
return df.repartitionByRange(fieldNum, org.apache.spark.sql.functions.col(zCols.get(0)));
|
||||
}
|
||||
Map<Integer, StructField> fieldMap = zCols
|
||||
.stream().collect(Collectors.toMap(e -> Arrays.asList(df.schema().fields()).indexOf(columnsMap.get(e)), e -> columnsMap.get(e)));
|
||||
// z-sort
|
||||
JavaRDD<Row> sortedRdd = df.toJavaRDD().map(row -> {
|
||||
List<byte[]> zBytesList = fieldMap.entrySet().stream().map(entry -> {
|
||||
int index = entry.getKey();
|
||||
StructField field = entry.getValue();
|
||||
DataType dataType = field.dataType();
|
||||
if (dataType instanceof LongType) {
|
||||
return ZOrderingUtil.longTo8Byte(row.isNullAt(index) ? Long.MAX_VALUE : row.getLong(index));
|
||||
} else if (dataType instanceof DoubleType) {
|
||||
return ZOrderingUtil.doubleTo8Byte(row.isNullAt(index) ? Double.MAX_VALUE : row.getDouble(index));
|
||||
} else if (dataType instanceof IntegerType) {
|
||||
return ZOrderingUtil.intTo8Byte(row.isNullAt(index) ? Integer.MAX_VALUE : row.getInt(index));
|
||||
} else if (dataType instanceof FloatType) {
|
||||
return ZOrderingUtil.doubleTo8Byte(row.isNullAt(index) ? Float.MAX_VALUE : row.getFloat(index));
|
||||
} else if (dataType instanceof StringType) {
|
||||
return ZOrderingUtil.utf8To8Byte(row.isNullAt(index) ? "" : row.getString(index));
|
||||
} else if (dataType instanceof DateType) {
|
||||
return ZOrderingUtil.longTo8Byte(row.isNullAt(index) ? Long.MAX_VALUE : row.getDate(index).getTime());
|
||||
} else if (dataType instanceof TimestampType) {
|
||||
return ZOrderingUtil.longTo8Byte(row.isNullAt(index) ? Long.MAX_VALUE : row.getTimestamp(index).getTime());
|
||||
} else if (dataType instanceof ByteType) {
|
||||
return ZOrderingUtil.byteTo8Byte(row.isNullAt(index) ? Byte.MAX_VALUE : row.getByte(index));
|
||||
} else if (dataType instanceof ShortType) {
|
||||
return ZOrderingUtil.intTo8Byte(row.isNullAt(index) ? Short.MAX_VALUE : row.getShort(index));
|
||||
} else if (dataType instanceof DecimalType) {
|
||||
return ZOrderingUtil.longTo8Byte(row.isNullAt(index) ? Long.MAX_VALUE : row.getDecimal(index).longValue());
|
||||
} else if (dataType instanceof BooleanType) {
|
||||
boolean value = row.isNullAt(index) ? false : row.getBoolean(index);
|
||||
return ZOrderingUtil.intTo8Byte(value ? 1 : 0);
|
||||
} else if (dataType instanceof BinaryType) {
|
||||
return ZOrderingUtil.paddingTo8Byte(row.isNullAt(index) ? new byte[] {0} : (byte[]) row.get(index));
|
||||
}
|
||||
return null;
|
||||
}).filter(f -> f != null).collect(Collectors.toList());
|
||||
byte[][] zBytes = new byte[zBytesList.size()][];
|
||||
for (int i = 0; i < zBytesList.size(); i++) {
|
||||
zBytes[i] = zBytesList.get(i);
|
||||
}
|
||||
List<Object> zVaules = new ArrayList<>();
|
||||
zVaules.addAll(scala.collection.JavaConverters.bufferAsJavaListConverter(row.toSeq().toBuffer()).asJava());
|
||||
zVaules.add(ZOrderingUtil.interleaving(zBytes, 8));
|
||||
return Row$.MODULE$.apply(JavaConversions.asScalaBuffer(zVaules));
|
||||
}).sortBy(f -> new ZorderingBinarySort((byte[]) f.get(fieldNum)), true, fileNum);
|
||||
|
||||
// create new StructType
|
||||
List<StructField> newFields = new ArrayList<>();
|
||||
newFields.addAll(Arrays.asList(df.schema().fields()));
|
||||
newFields.add(new StructField("zIndex", BinaryType$.MODULE$, true, Metadata.empty()));
|
||||
|
||||
// create new DataFrame
|
||||
return df.sparkSession().createDataFrame(sortedRdd, StructType$.MODULE$.apply(newFields)).drop("zIndex");
|
||||
}
|
||||
|
||||
public static Dataset<Row> createZIndexedDataFrameByMapValue(Dataset<Row> df, String zCols, int fileNum) {
|
||||
if (zCols == null || zCols.isEmpty() || fileNum <= 0) {
|
||||
return df;
|
||||
}
|
||||
return createZIndexedDataFrameByMapValue(df,
|
||||
Arrays.stream(zCols.split(",")).map(f -> f.trim()).collect(Collectors.toList()), fileNum);
|
||||
}
|
||||
|
||||
public static Dataset<Row> createZIndexedDataFrameBySample(Dataset<Row> df, List<String> zCols, int fileNum) {
|
||||
return RangeSampleSort$.MODULE$.sortDataFrameBySample(df, JavaConversions.asScalaBuffer(zCols), fileNum);
|
||||
}
|
||||
|
||||
public static Dataset<Row> createZIndexedDataFrameBySample(Dataset<Row> df, String zCols, int fileNum) {
|
||||
if (zCols == null || zCols.isEmpty() || fileNum <= 0) {
|
||||
return df;
|
||||
}
|
||||
return createZIndexedDataFrameBySample(df, Arrays.stream(zCols.split(",")).map(f -> f.trim()).collect(Collectors.toList()), fileNum);
|
||||
}
|
||||
|
||||
/**
|
||||
* Parse min/max statistics stored in parquet footers for z-sort cols.
|
||||
* no support collect statistics from timeStampType, since parquet file has not collect the statistics for timeStampType.
|
||||
* to do adapt for rfc-27
|
||||
*
|
||||
* @param df a spark DataFrame holds parquet files to be read.
|
||||
* @param cols z-sort cols
|
||||
* @return a dataFrame holds all statistics info.
|
||||
*/
|
||||
public static Dataset<Row> getMinMaxValue(Dataset<Row> df, List<String> cols) {
|
||||
Map<String, DataType> columnsMap = Arrays.stream(df.schema().fields()).collect(Collectors.toMap(e -> e.name(), e -> e.dataType()));
|
||||
|
||||
List<String> scanFiles = Arrays.asList(df.inputFiles());
|
||||
SparkContext sc = df.sparkSession().sparkContext();
|
||||
JavaSparkContext jsc = new JavaSparkContext(sc);
|
||||
|
||||
SerializableConfiguration serializableConfiguration = new SerializableConfiguration(sc.hadoopConfiguration());
|
||||
int numParallelism = (scanFiles.size() / 3 + 1);
|
||||
List<HoodieColumnRangeMetadata<Comparable>> colMinMaxInfos = new ArrayList<>();
|
||||
String previousJobDescription = sc.getLocalProperty(SPARK_JOB_DESCRIPTION);
|
||||
try {
|
||||
String description = "Listing parquet column statistics";
|
||||
jsc.setJobDescription(description);
|
||||
colMinMaxInfos = jsc.parallelize(scanFiles, numParallelism).mapPartitions(paths -> {
|
||||
Configuration conf = serializableConfiguration.value();
|
||||
ParquetUtils parquetUtils = (ParquetUtils) BaseFileUtils.getInstance(HoodieFileFormat.PARQUET);
|
||||
List<Collection<HoodieColumnRangeMetadata<Comparable>>> results = new ArrayList<>();
|
||||
while (paths.hasNext()) {
|
||||
String path = paths.next();
|
||||
results.add(parquetUtils.readRangeFromParquetMetadata(conf, new Path(path), cols));
|
||||
}
|
||||
return results.stream().flatMap(f -> f.stream()).iterator();
|
||||
}).collect();
|
||||
} finally {
|
||||
jsc.setJobDescription(previousJobDescription);
|
||||
}
|
||||
|
||||
Map<String, List<HoodieColumnRangeMetadata<Comparable>>> fileToStatsListMap = colMinMaxInfos.stream().collect(Collectors.groupingBy(e -> e.getFilePath()));
|
||||
JavaRDD<Row> allMetaDataRDD = jsc.parallelize(fileToStatsListMap.values().stream().collect(Collectors.toList()), 1).map(f -> {
|
||||
int colSize = f.size();
|
||||
if (colSize == 0) {
|
||||
return null;
|
||||
} else {
|
||||
List<Object> rows = new ArrayList<>();
|
||||
rows.add(f.get(0).getFilePath());
|
||||
cols.stream().forEach(col -> {
|
||||
HoodieColumnRangeMetadata<Comparable> currentColRangeMetaData =
|
||||
f.stream().filter(s -> s.getColumnName().trim().equalsIgnoreCase(col)).findFirst().orElse(null);
|
||||
DataType colType = columnsMap.get(col);
|
||||
if (currentColRangeMetaData == null || colType == null) {
|
||||
throw new HoodieException(String.format("cannot collect min/max statistics for col: %s", col));
|
||||
}
|
||||
if (colType instanceof IntegerType) {
|
||||
rows.add(currentColRangeMetaData.getMinValue());
|
||||
rows.add(currentColRangeMetaData.getMaxValue());
|
||||
} else if (colType instanceof DoubleType) {
|
||||
rows.add(currentColRangeMetaData.getMinValue());
|
||||
rows.add(currentColRangeMetaData.getMaxValue());
|
||||
} else if (colType instanceof StringType) {
|
||||
String minString = new String(((Binary)currentColRangeMetaData.getMinValue()).getBytes());
|
||||
String maxString = new String(((Binary)currentColRangeMetaData.getMaxValue()).getBytes());
|
||||
rows.add(minString);
|
||||
rows.add(maxString);
|
||||
} else if (colType instanceof DecimalType) {
|
||||
Double minDecimal = Double.parseDouble(currentColRangeMetaData.getStringifier().stringify(Long.valueOf(currentColRangeMetaData.getMinValue().toString())));
|
||||
Double maxDecimal = Double.parseDouble(currentColRangeMetaData.getStringifier().stringify(Long.valueOf(currentColRangeMetaData.getMaxValue().toString())));
|
||||
rows.add(BigDecimal.valueOf(minDecimal));
|
||||
rows.add(BigDecimal.valueOf(maxDecimal));
|
||||
} else if (colType instanceof DateType) {
|
||||
rows.add(java.sql.Date.valueOf(currentColRangeMetaData.getStringifier().stringify((int)currentColRangeMetaData.getMinValue())));
|
||||
rows.add(java.sql.Date.valueOf(currentColRangeMetaData.getStringifier().stringify((int)currentColRangeMetaData.getMaxValue())));
|
||||
} else if (colType instanceof LongType) {
|
||||
rows.add(currentColRangeMetaData.getMinValue());
|
||||
rows.add(currentColRangeMetaData.getMaxValue());
|
||||
} else if (colType instanceof ShortType) {
|
||||
rows.add(Short.parseShort(currentColRangeMetaData.getMinValue().toString()));
|
||||
rows.add(Short.parseShort(currentColRangeMetaData.getMaxValue().toString()));
|
||||
} else if (colType instanceof FloatType) {
|
||||
rows.add(currentColRangeMetaData.getMinValue());
|
||||
rows.add(currentColRangeMetaData.getMaxValue());
|
||||
} else if (colType instanceof BinaryType) {
|
||||
rows.add(((Binary)currentColRangeMetaData.getMinValue()).getBytes());
|
||||
rows.add(((Binary)currentColRangeMetaData.getMaxValue()).getBytes());
|
||||
} else if (colType instanceof BooleanType) {
|
||||
rows.add(currentColRangeMetaData.getMinValue());
|
||||
rows.add(currentColRangeMetaData.getMaxValue());
|
||||
} else if (colType instanceof ByteType) {
|
||||
rows.add(Byte.valueOf(currentColRangeMetaData.getMinValue().toString()));
|
||||
rows.add(Byte.valueOf(currentColRangeMetaData.getMaxValue().toString()));
|
||||
} else {
|
||||
throw new HoodieException(String.format("Not support type: %s", colType));
|
||||
}
|
||||
rows.add(currentColRangeMetaData.getNumNulls());
|
||||
});
|
||||
return Row$.MODULE$.apply(JavaConversions.asScalaBuffer(rows));
|
||||
}
|
||||
}).filter(f -> f != null);
|
||||
List<StructField> allMetaDataSchema = new ArrayList<>();
|
||||
allMetaDataSchema.add(new StructField("file", StringType$.MODULE$, true, Metadata.empty()));
|
||||
cols.forEach(col -> {
|
||||
allMetaDataSchema.add(new StructField(col + "_minValue", columnsMap.get(col), true, Metadata.empty()));
|
||||
allMetaDataSchema.add(new StructField(col + "_maxValue", columnsMap.get(col), true, Metadata.empty()));
|
||||
allMetaDataSchema.add(new StructField(col + "_num_nulls", LongType$.MODULE$, true, Metadata.empty()));
|
||||
});
|
||||
return df.sparkSession().createDataFrame(allMetaDataRDD, StructType$.MODULE$.apply(allMetaDataSchema));
|
||||
}
|
||||
|
||||
public static Dataset<Row> getMinMaxValue(Dataset<Row> df, String cols) {
|
||||
List<String> rawCols = Arrays.asList(cols.split(",")).stream().map(f -> f.trim()).collect(Collectors.toList());
|
||||
return getMinMaxValue(df, rawCols);
|
||||
}
|
||||
|
||||
/**
|
||||
* Update statistics info.
|
||||
* this method will update old index table by full out join,
|
||||
* and save the updated table into a new index table based on commitTime.
|
||||
* old index table will be cleaned also.
|
||||
*
|
||||
* @param df a spark DataFrame holds parquet files to be read.
|
||||
* @param cols z-sort cols.
|
||||
* @param indexPath index store path.
|
||||
* @param commitTime current operation commitTime.
|
||||
* @param validateCommits all validate commits for current table.
|
||||
* @return
|
||||
*/
|
||||
public static void saveStatisticsInfo(Dataset<Row> df, String cols, String indexPath, String commitTime, List<String> validateCommits) {
|
||||
Path savePath = new Path(indexPath, commitTime);
|
||||
SparkSession spark = df.sparkSession();
|
||||
FileSystem fs = FSUtils.getFs(indexPath, spark.sparkContext().hadoopConfiguration());
|
||||
Dataset<Row> statisticsDF = ZCurveOptimizeHelper.getMinMaxValue(df, cols);
|
||||
// try to find last validate index table from index path
|
||||
try {
|
||||
if (fs.exists(new Path(indexPath))) {
|
||||
List<String> allIndexTables = Arrays
|
||||
.stream(fs.listStatus(new Path(indexPath))).filter(f -> f.isDirectory()).map(f -> f.getPath().getName()).collect(Collectors.toList());
|
||||
List<String> candidateIndexTables = allIndexTables.stream().filter(f -> validateCommits.contains(f)).sorted().collect(Collectors.toList());
|
||||
List<String> residualTables = allIndexTables.stream().filter(f -> !validateCommits.contains(f)).collect(Collectors.toList());
|
||||
Option<Dataset> latestIndexData = Option.empty();
|
||||
if (!candidateIndexTables.isEmpty()) {
|
||||
latestIndexData = Option.of(spark.read().load(new Path(indexPath, candidateIndexTables.get(candidateIndexTables.size() - 1)).toString()));
|
||||
// clean old index table, keep at most 1 index table.
|
||||
candidateIndexTables.remove(candidateIndexTables.size() - 1);
|
||||
candidateIndexTables.forEach(f -> {
|
||||
try {
|
||||
fs.delete(new Path(indexPath, f));
|
||||
} catch (IOException ie) {
|
||||
throw new HoodieException(ie);
|
||||
}
|
||||
});
|
||||
}
|
||||
|
||||
// clean residualTables
|
||||
// retried cluster operations at the same instant time is also considered,
|
||||
// the residual files produced by retried are cleaned up before save statistics
|
||||
// save statistics info to index table which named commitTime
|
||||
residualTables.forEach(f -> {
|
||||
try {
|
||||
fs.delete(new Path(indexPath, f));
|
||||
} catch (IOException ie) {
|
||||
throw new HoodieException(ie);
|
||||
}
|
||||
});
|
||||
|
||||
if (latestIndexData.isPresent() && latestIndexData.get().schema().equals(statisticsDF.schema())) {
|
||||
// update the statistics info
|
||||
String originalTable = "indexTable_" + java.util.UUID.randomUUID().toString().replace("-", "");
|
||||
String updateTable = "updateTable_" + java.util.UUID.randomUUID().toString().replace("-", "");
|
||||
latestIndexData.get().registerTempTable(originalTable);
|
||||
statisticsDF.registerTempTable(updateTable);
|
||||
// update table by full out join
|
||||
List columns = Arrays.asList(statisticsDF.schema().fieldNames());
|
||||
spark.sql(HoodieSparkUtils$
|
||||
.MODULE$.createMergeSql(originalTable, updateTable, JavaConversions.asScalaBuffer(columns))).repartition(1).write().save(savePath.toString());
|
||||
}
|
||||
} else {
|
||||
statisticsDF.repartition(1).write().mode("overwrite").save(savePath.toString());
|
||||
}
|
||||
} catch (IOException e) {
|
||||
throw new HoodieException(e);
|
||||
}
|
||||
}
|
||||
}
|
||||
@@ -19,6 +19,7 @@
|
||||
package org.apache.hudi
|
||||
|
||||
import java.util.Properties
|
||||
|
||||
import org.apache.avro.Schema
|
||||
import org.apache.avro.generic.GenericRecord
|
||||
import org.apache.hadoop.fs.{FileSystem, Path}
|
||||
@@ -35,6 +36,7 @@ import org.apache.spark.sql.avro.SchemaConverters
|
||||
import org.apache.spark.sql.catalyst.encoders.RowEncoder
|
||||
import org.apache.spark.sql.catalyst.expressions.{AttributeReference, Expression, Literal}
|
||||
import org.apache.spark.sql.execution.datasources.{FileStatusCache, InMemoryFileIndex}
|
||||
import org.apache.spark.sql.functions._
|
||||
import org.apache.spark.sql.sources._
|
||||
import org.apache.spark.sql.types.{StringType, StructField, StructType}
|
||||
import org.apache.spark.sql.{DataFrame, SparkSession}
|
||||
@@ -283,4 +285,43 @@ object HoodieSparkUtils extends SparkAdapterSupport {
|
||||
s"${tableSchema.fieldNames.mkString(",")}")
|
||||
AttributeReference(columnName, field.get.dataType, field.get.nullable)()
|
||||
}
|
||||
|
||||
/**
|
||||
* Create merge sql to merge leftTable and right table.
|
||||
*
|
||||
* @param leftTable table name.
|
||||
* @param rightTable table name.
|
||||
* @param cols merged cols.
|
||||
* @return merge sql.
|
||||
*/
|
||||
def createMergeSql(leftTable: String, rightTable: String, cols: Seq[String]): String = {
|
||||
var selectsql = ""
|
||||
for (i <- (0 to cols.size-1)) {
|
||||
selectsql = selectsql + s" if (${leftTable}.${cols(0)} is null, ${rightTable}.${cols(i)}, ${leftTable}.${cols(i)}) as ${cols(i)} ,"
|
||||
}
|
||||
"select " + selectsql.dropRight(1) + s" from ${leftTable} full join ${rightTable} on ${leftTable}.${cols(0)} = ${rightTable}.${cols(0)}"
|
||||
}
|
||||
|
||||
/**
|
||||
* Collect min/max statistics for candidate cols.
|
||||
* support all col types.
|
||||
*
|
||||
* @param df dataFrame holds read files.
|
||||
* @param cols candidate cols to collect statistics.
|
||||
* @return
|
||||
*/
|
||||
def getMinMaxValueSpark(df: DataFrame, cols: Seq[String]): DataFrame = {
|
||||
val sqlContext = df.sparkSession.sqlContext
|
||||
import sqlContext.implicits._
|
||||
|
||||
val values = cols.flatMap(c => Seq( min(col(c)).as(c + "_minValue"), max(col(c)).as(c + "_maxValue"), count(c).as(c + "_noNullCount")))
|
||||
val valueCounts = count("*").as("totalNum")
|
||||
val projectValues = Seq(col("file")) ++ cols.flatMap(c =>
|
||||
Seq(col(c + "_minValue"), col(c + "_maxValue"), expr(s"totalNum - ${c + "_noNullCount"}").as(c + "_num_nulls")))
|
||||
|
||||
val result = df.select(input_file_name() as "file", col("*"))
|
||||
.groupBy($"file")
|
||||
.agg(valueCounts, values: _*).select(projectValues:_*)
|
||||
result
|
||||
}
|
||||
}
|
||||
|
||||
@@ -0,0 +1,526 @@
|
||||
/*
|
||||
* 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.spark.sql.hudi.execution
|
||||
|
||||
import java.util
|
||||
|
||||
import org.apache.hudi.config.HoodieClusteringConfig
|
||||
import org.apache.spark.rdd.{PartitionPruningRDD, RDD}
|
||||
import org.apache.spark.sql.catalyst.expressions.{Ascending, Attribute, BoundReference, SortOrder, UnsafeProjection, UnsafeRow}
|
||||
import org.apache.hudi.optimize.ZOrderingUtil
|
||||
import org.apache.spark.sql.{DataFrame, Row}
|
||||
import org.apache.spark.sql.catalyst.InternalRow
|
||||
import org.apache.spark.sql.catalyst.expressions.codegen.LazilyGeneratedOrdering
|
||||
import org.apache.spark.sql.types._
|
||||
import org.apache.spark.util.MutablePair
|
||||
import org.apache.spark.util.random.SamplingUtils
|
||||
|
||||
import scala.collection.mutable
|
||||
import scala.collection.mutable.ArrayBuffer
|
||||
import scala.reflect.{ClassTag, classTag}
|
||||
import scala.util.hashing.byteswap32
|
||||
|
||||
class RangeSample[K: ClassTag, V](
|
||||
zEncodeNum: Int,
|
||||
rdd: RDD[_ <: Product2[K, V]],
|
||||
private var ascend: Boolean = true,
|
||||
val samplePointsPerPartitionHint: Int = 20) extends Serializable {
|
||||
|
||||
// We allow zEncodeNum = 0, which happens when sorting an empty RDD under the default settings.
|
||||
require(zEncodeNum >= 0, s"Number of zEncodeNum cannot be negative but found $zEncodeNum.")
|
||||
require(samplePointsPerPartitionHint > 0,
|
||||
s"Sample points per partition must be greater than 0 but found $samplePointsPerPartitionHint")
|
||||
|
||||
def getRangeBounds(): ArrayBuffer[(K, Float)] = {
|
||||
if (zEncodeNum <= 1) {
|
||||
ArrayBuffer.empty[(K, Float)]
|
||||
} else {
|
||||
// This is the sample size we need to have roughly balanced output partitions, capped at 1M.
|
||||
// Cast to double to avoid overflowing ints or longs
|
||||
val sampleSize = math.min(samplePointsPerPartitionHint.toDouble * zEncodeNum, 1e6)
|
||||
// Assume the input partitions are roughly balanced and over-sample a little bit.
|
||||
val sampleSizePerPartition = math.ceil(3.0 * sampleSize / rdd.partitions.length).toInt
|
||||
val (numItems, sketched) = sketch(rdd.map(_._1), sampleSizePerPartition)
|
||||
if (numItems == 0L) {
|
||||
ArrayBuffer.empty[(K, Float)]
|
||||
} else {
|
||||
// If a partition contains much more than the average number of items, we re-sample from it
|
||||
// to ensure that enough items are collected from that partition.
|
||||
val fraction = math.min(sampleSize / math.max(numItems, 1L), 1.0)
|
||||
val candidates = ArrayBuffer.empty[(K, Float)]
|
||||
val imbalancedPartitions = mutable.Set.empty[Int]
|
||||
|
||||
sketched.foreach { case (idx, n, sample) =>
|
||||
if (fraction * n > sampleSizePerPartition) {
|
||||
imbalancedPartitions += idx
|
||||
} else {
|
||||
// The weight is 1 over the sampling probability.
|
||||
val weight = (n.toDouble / sample.length).toFloat
|
||||
for (key <- sample) {
|
||||
candidates += ((key, weight))
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
if (imbalancedPartitions.nonEmpty) {
|
||||
// Re-sample imbalanced partitions with the desired sampling probability.
|
||||
val imbalanced = new PartitionPruningRDD(rdd.map(_._1), imbalancedPartitions.contains)
|
||||
val seed = byteswap32(-rdd.id - 1)
|
||||
val reSampled = imbalanced.sample(withReplacement = false, fraction, seed).collect()
|
||||
val weight = (1.0 / fraction).toFloat
|
||||
candidates ++= reSampled.map(x => (x, weight))
|
||||
}
|
||||
candidates
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* Determines the bounds for range partitioning from candidates with weights indicating how many
|
||||
* items each represents. Usually this is 1 over the probability used to sample this candidate.
|
||||
*
|
||||
* @param candidates unordered candidates with weights
|
||||
* @param partitions number of partitions
|
||||
* @return selected bounds
|
||||
*/
|
||||
def determineBound[K : Ordering : ClassTag](
|
||||
candidates: ArrayBuffer[(K, Float)],
|
||||
partitions: Int, ordering: Ordering[K]): Array[K] = {
|
||||
val ordered = candidates.sortBy(_._1)(ordering)
|
||||
val numCandidates = ordered.size
|
||||
val sumWeights = ordered.map(_._2.toDouble).sum
|
||||
val step = sumWeights / partitions
|
||||
var cumWeight = 0.0
|
||||
var target = step
|
||||
val bounds = ArrayBuffer.empty[K]
|
||||
var i = 0
|
||||
var j = 0
|
||||
var previousBound = Option.empty[K]
|
||||
while ((i < numCandidates) && (j < partitions - 1)) {
|
||||
val (key, weight) = ordered(i)
|
||||
cumWeight += weight
|
||||
if (cumWeight >= target) {
|
||||
// Skip duplicate values.
|
||||
if (previousBound.isEmpty || ordering.gt(key, previousBound.get)) {
|
||||
bounds += key
|
||||
target += step
|
||||
j += 1
|
||||
previousBound = Some(key)
|
||||
}
|
||||
}
|
||||
i += 1
|
||||
}
|
||||
bounds.toArray
|
||||
}
|
||||
|
||||
def determineRowBounds[K : Ordering : ClassTag](
|
||||
candidates: ArrayBuffer[(K, Float)],
|
||||
partitions: Int, orderings: Seq[Ordering[K]],
|
||||
attributes: Seq[Attribute]): Array[Array[UnsafeRow]] = {
|
||||
|
||||
orderings.zipWithIndex.map { case (ordering, index) =>
|
||||
val ordered = candidates.sortBy(_._1)(ordering)
|
||||
val numCandidates = ordered.size
|
||||
val sumWeights = ordered.map(_._2.toDouble).sum
|
||||
val step = sumWeights / partitions
|
||||
var cumWeight = 0.0
|
||||
var target = step
|
||||
val bounds = ArrayBuffer.empty[K]
|
||||
var i = 0
|
||||
var j = 0
|
||||
var previousBound = Option.empty[K]
|
||||
while ((i < numCandidates) && (j < partitions - 1)) {
|
||||
val (key, weight) = ordered(i)
|
||||
cumWeight += weight
|
||||
if (cumWeight >= target) {
|
||||
// Skip duplicate values.
|
||||
if (previousBound.isEmpty || ordering.gt(key, previousBound.get)) {
|
||||
bounds += key
|
||||
target += step
|
||||
j += 1
|
||||
previousBound = Some(key)
|
||||
}
|
||||
}
|
||||
i += 1
|
||||
}
|
||||
// build project
|
||||
val project = UnsafeProjection.create(Seq(attributes(index)), attributes)
|
||||
bounds.map { bound =>
|
||||
val row = bound.asInstanceOf[UnsafeRow]
|
||||
project(row).copy()
|
||||
}.toArray
|
||||
}.toArray
|
||||
}
|
||||
|
||||
/**
|
||||
* Sketches the input RDD via reservoir sampling on each partition.
|
||||
*
|
||||
* @param rdd the input RDD to sketch
|
||||
* @param sampleSizePerPartition max sample size per partition
|
||||
* @return (total number of items, an array of (partitionId, number of items, sample))
|
||||
*/
|
||||
def sketch[K: ClassTag](
|
||||
rdd: RDD[K],
|
||||
sampleSizePerPartition: Int): (Long, Array[(Int, Long, Array[K])]) = {
|
||||
val shift = rdd.id
|
||||
// val classTagK = classTag[K] // to avoid serializing the entire partitioner object
|
||||
val sketched = rdd.mapPartitionsWithIndex { (idx, iter) =>
|
||||
val seed = byteswap32(idx ^ (shift << 16))
|
||||
val (sample, n) = SamplingUtils.reservoirSampleAndCount(
|
||||
iter, sampleSizePerPartition, seed)
|
||||
Iterator((idx, n, sample))
|
||||
}.collect()
|
||||
val numItems = sketched.map(_._2).sum
|
||||
(numItems, sketched)
|
||||
}
|
||||
}
|
||||
|
||||
class RawDecisionBound[K : Ordering : ClassTag](ordering: Ordering[K]) extends Serializable {
|
||||
|
||||
private var binarySearch: ((Array[K], K) => Int) = {
|
||||
// For primitive keys, we can use the natural ordering. Otherwise, use the Ordering comparator.
|
||||
classTag[K] match {
|
||||
case ClassTag.Float =>
|
||||
(l, x) => util.Arrays.binarySearch(l.asInstanceOf[Array[Float]], x.asInstanceOf[Float])
|
||||
case ClassTag.Double =>
|
||||
(l, x) => util.Arrays.binarySearch(l.asInstanceOf[Array[Double]], x.asInstanceOf[Double])
|
||||
case ClassTag.Byte =>
|
||||
(l, x) => util.Arrays.binarySearch(l.asInstanceOf[Array[Byte]], x.asInstanceOf[Byte])
|
||||
case ClassTag.Char =>
|
||||
(l, x) => util.Arrays.binarySearch(l.asInstanceOf[Array[Char]], x.asInstanceOf[Char])
|
||||
case ClassTag.Short =>
|
||||
(l, x) => util.Arrays.binarySearch(l.asInstanceOf[Array[Short]], x.asInstanceOf[Short])
|
||||
case ClassTag.Int =>
|
||||
(l, x) => util.Arrays.binarySearch(l.asInstanceOf[Array[Int]], x.asInstanceOf[Int])
|
||||
case ClassTag.Long =>
|
||||
(l, x) => util.Arrays.binarySearch(l.asInstanceOf[Array[Long]], x.asInstanceOf[Long])
|
||||
case _ =>
|
||||
val comparator = ordering.asInstanceOf[java.util.Comparator[Any]]
|
||||
(l, x) => util.Arrays.binarySearch(l.asInstanceOf[Array[AnyRef]], x, comparator)
|
||||
}
|
||||
}
|
||||
|
||||
def getBound(key: Any, candidateBounds: Array[K]): Int = {
|
||||
val k = key.asInstanceOf[K]
|
||||
var bound = 0
|
||||
if (candidateBounds.length <= 128) {
|
||||
while(bound < candidateBounds.length && ordering.gt(k, candidateBounds(bound))) {
|
||||
bound += 1
|
||||
}
|
||||
} else {
|
||||
bound = binarySearch(candidateBounds, k)
|
||||
if (bound < 0 ) {
|
||||
bound = -bound - 1
|
||||
}
|
||||
if (bound > candidateBounds.length) {
|
||||
bound = candidateBounds.length
|
||||
}
|
||||
}
|
||||
bound
|
||||
}
|
||||
}
|
||||
|
||||
case class ZorderingBinarySort(b: Array[Byte]) extends Ordered[ZorderingBinarySort] with Serializable {
|
||||
override def compare(that: ZorderingBinarySort): Int = {
|
||||
val len = this.b.length
|
||||
ZOrderingUtil.compareTo(this.b, 0, len, that.b, 0, len)
|
||||
}
|
||||
}
|
||||
|
||||
object RangeSampleSort {
|
||||
|
||||
/**
|
||||
* create z-order DataFrame by sample
|
||||
* support all col types
|
||||
*/
|
||||
def sortDataFrameBySampleSupportAllTypes(df: DataFrame, zCols: Seq[String], fileNum: Int): DataFrame = {
|
||||
val spark = df.sparkSession
|
||||
val internalRdd = df.queryExecution.toRdd
|
||||
val schema = df.schema
|
||||
val outputAttributes = df.queryExecution.analyzed.output
|
||||
val sortingExpressions = outputAttributes.filter(p => zCols.contains(p.name))
|
||||
if (sortingExpressions.length == 0 || sortingExpressions.length != zCols.size) {
|
||||
df
|
||||
} else {
|
||||
val zOrderBounds = df.sparkSession.sessionState.conf.getConfString(
|
||||
HoodieClusteringConfig.LAYOUT_OPTIMIZE_BUILD_CURVE_SAMPLE_SIZE.key,
|
||||
HoodieClusteringConfig.LAYOUT_OPTIMIZE_BUILD_CURVE_SAMPLE_SIZE.defaultValue.toString).toInt
|
||||
|
||||
val sampleRdd = internalRdd.mapPartitionsInternal { iter =>
|
||||
val projection = UnsafeProjection.create(sortingExpressions, outputAttributes)
|
||||
val mutablePair = new MutablePair[InternalRow, Null]()
|
||||
// Internally, RangePartitioner runs a job on the RDD that samples keys to compute
|
||||
// partition bounds. To get accurate samples, we need to copy the mutable keys.
|
||||
iter.map(row => mutablePair.update(projection(row).copy(), null))
|
||||
}
|
||||
|
||||
val orderings = sortingExpressions.map(SortOrder(_, Ascending)).zipWithIndex.map { case (ord, i) =>
|
||||
ord.copy(child = BoundReference(i, ord.dataType, ord.nullable))
|
||||
}
|
||||
|
||||
val lazyGeneratedOrderings = orderings.map(ord => new LazilyGeneratedOrdering(Seq(ord)))
|
||||
|
||||
val sample = new RangeSample(zOrderBounds, sampleRdd)
|
||||
|
||||
val rangeBounds = sample.getRangeBounds()
|
||||
|
||||
implicit val ordering1 = lazyGeneratedOrderings(0)
|
||||
|
||||
val sampleBounds = sample.determineRowBounds(rangeBounds, math.min(zOrderBounds, rangeBounds.length), lazyGeneratedOrderings, sortingExpressions)
|
||||
|
||||
val origin_orderings = sortingExpressions.map(SortOrder(_, Ascending)).map { ord =>
|
||||
ord.copy(child = BoundReference(0, ord.dataType, ord.nullable))
|
||||
}
|
||||
|
||||
val origin_lazyGeneratedOrderings = origin_orderings.map(ord => new LazilyGeneratedOrdering(Seq(ord)))
|
||||
|
||||
// expand bounds.
|
||||
// maybe it's better to use the value of "spark.zorder.bounds.number" as maxLength,
|
||||
// however this will lead to extra time costs when all zorder cols distinct count values are less then "spark.zorder.bounds.number"
|
||||
val maxLength = sampleBounds.map(_.length).max
|
||||
val expandSampleBoundsWithFactor = sampleBounds.map { bound =>
|
||||
val fillFactor = maxLength / bound.size.toDouble
|
||||
(bound, fillFactor)
|
||||
}
|
||||
|
||||
val boundBroadCast = spark.sparkContext.broadcast(expandSampleBoundsWithFactor)
|
||||
|
||||
val indexRdd = internalRdd.mapPartitionsInternal { iter =>
|
||||
val boundsWithFactor = boundBroadCast.value
|
||||
import java.util.concurrent.ThreadLocalRandom
|
||||
val threadLocalRandom = ThreadLocalRandom.current
|
||||
val maxBoundNum = boundsWithFactor.map(_._1.length).max
|
||||
val origin_Projections = sortingExpressions.map { se =>
|
||||
UnsafeProjection.create(Seq(se), outputAttributes)
|
||||
}
|
||||
|
||||
iter.map { unsafeRow =>
|
||||
val interleaveValues = origin_Projections.zip(origin_lazyGeneratedOrderings).zipWithIndex.map { case ((rowProject, lazyOrdering), index) =>
|
||||
val row = rowProject(unsafeRow)
|
||||
val decisionBound = new RawDecisionBound(lazyOrdering)
|
||||
if (row.isNullAt(0)) {
|
||||
maxBoundNum + 1
|
||||
} else {
|
||||
val (bound, factor) = boundsWithFactor(index)
|
||||
if (factor > 1) {
|
||||
val currentRank = decisionBound.getBound(row, bound.asInstanceOf[Array[InternalRow]])
|
||||
currentRank*factor.toInt + threadLocalRandom.nextInt(factor.toInt)
|
||||
} else {
|
||||
decisionBound.getBound(row, bound.asInstanceOf[Array[InternalRow]])
|
||||
}
|
||||
}
|
||||
}.toArray.map(ZOrderingUtil.intTo8Byte(_))
|
||||
val zValues = ZOrderingUtil.interleaving(interleaveValues, 8)
|
||||
val mutablePair = new MutablePair[InternalRow, Array[Byte]]()
|
||||
|
||||
mutablePair.update(unsafeRow, zValues)
|
||||
}
|
||||
}.sortBy(x => ZorderingBinarySort(x._2), numPartitions = fileNum).map(_._1)
|
||||
spark.internalCreateDataFrame(indexRdd, schema)
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* create z-order DataFrame by sample
|
||||
* first, sample origin data to get z-cols bounds, then create z-order DataFrame
|
||||
* support all type data.
|
||||
* this method need more resource and cost more time than createZIndexedDataFrameByMapValue
|
||||
*/
|
||||
def sortDataFrameBySample(df: DataFrame, zCols: Seq[String], fileNum: Int): DataFrame = {
|
||||
val spark = df.sparkSession
|
||||
val columnsMap = df.schema.fields.map(item => (item.name, item)).toMap
|
||||
val fieldNum = df.schema.fields.length
|
||||
val checkCols = zCols.filter(col => columnsMap(col) != null)
|
||||
|
||||
if (zCols.isEmpty || checkCols.isEmpty) {
|
||||
df
|
||||
} else {
|
||||
val zFields = zCols.map { col =>
|
||||
val newCol = columnsMap(col)
|
||||
if (newCol == null) {
|
||||
(-1, null)
|
||||
} else {
|
||||
newCol.dataType match {
|
||||
case LongType | DoubleType | FloatType | StringType | IntegerType | DateType | TimestampType | ShortType | ByteType =>
|
||||
(df.schema.fields.indexOf(newCol), newCol)
|
||||
case d: DecimalType =>
|
||||
(df.schema.fields.indexOf(newCol), newCol)
|
||||
case _ =>
|
||||
(-1, null)
|
||||
}
|
||||
}
|
||||
}.filter(_._1 != -1)
|
||||
// Complex type found, use createZIndexedDataFrameByRange
|
||||
if (zFields.length != zCols.length) {
|
||||
return sortDataFrameBySampleSupportAllTypes(df, zCols, fieldNum)
|
||||
}
|
||||
|
||||
val rawRdd = df.rdd
|
||||
val sampleRdd = rawRdd.map { row =>
|
||||
val values = zFields.map { case (index, field) =>
|
||||
field.dataType match {
|
||||
case LongType =>
|
||||
if (row.isNullAt(index)) Long.MaxValue else row.getLong(index)
|
||||
case DoubleType =>
|
||||
if (row.isNullAt(index)) Long.MaxValue else java.lang.Double.doubleToLongBits(row.getDouble(index))
|
||||
case IntegerType =>
|
||||
if (row.isNullAt(index)) Long.MaxValue else row.getInt(index).toLong
|
||||
case FloatType =>
|
||||
if (row.isNullAt(index)) Long.MaxValue else java.lang.Double.doubleToLongBits(row.getFloat(index).toDouble)
|
||||
case StringType =>
|
||||
if (row.isNullAt(index)) "" else row.getString(index)
|
||||
case DateType =>
|
||||
if (row.isNullAt(index)) Long.MaxValue else row.getDate(index).getTime
|
||||
case TimestampType =>
|
||||
if (row.isNullAt(index)) Long.MaxValue else row.getTimestamp(index).getTime
|
||||
case ByteType =>
|
||||
if (row.isNullAt(index)) Long.MaxValue else row.getByte(index).toLong
|
||||
case ShortType =>
|
||||
if (row.isNullAt(index)) Long.MaxValue else row.getShort(index).toLong
|
||||
case d: DecimalType =>
|
||||
if (row.isNullAt(index)) Long.MaxValue else row.getDecimal(index).longValue()
|
||||
case _ =>
|
||||
null
|
||||
}
|
||||
}.filter(v => v != null).toArray
|
||||
(values, null)
|
||||
}
|
||||
val zOrderBounds = df.sparkSession.sessionState.conf.getConfString(
|
||||
HoodieClusteringConfig.LAYOUT_OPTIMIZE_BUILD_CURVE_SAMPLE_SIZE.key,
|
||||
HoodieClusteringConfig.LAYOUT_OPTIMIZE_BUILD_CURVE_SAMPLE_SIZE.defaultValue.toString).toInt
|
||||
val sample = new RangeSample(zOrderBounds, sampleRdd)
|
||||
val rangeBounds = sample.getRangeBounds()
|
||||
val sampleBounds = {
|
||||
val candidateColNumber = rangeBounds.head._1.length
|
||||
(0 to candidateColNumber - 1).map { i =>
|
||||
val colRangeBound = rangeBounds.map(x => (x._1(i), x._2))
|
||||
|
||||
if (colRangeBound.head._1.isInstanceOf[String]) {
|
||||
sample.determineBound(colRangeBound.asInstanceOf[ArrayBuffer[(String, Float)]], math.min(zOrderBounds, rangeBounds.length), Ordering[String])
|
||||
} else {
|
||||
sample.determineBound(colRangeBound.asInstanceOf[ArrayBuffer[(Long, Float)]], math.min(zOrderBounds, rangeBounds.length), Ordering[Long])
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// expand bounds.
|
||||
// maybe it's better to use the value of "spark.zorder.bounds.number" as maxLength,
|
||||
// however this will lead to extra time costs when all zorder cols distinct count values are less then "spark.zorder.bounds.number"
|
||||
val maxLength = sampleBounds.map(_.length).max
|
||||
val expandSampleBoundsWithFactor = sampleBounds.map { bound =>
|
||||
val fillFactor = maxLength / bound.size
|
||||
val newBound = new Array[Double](bound.length * fillFactor)
|
||||
if (bound.isInstanceOf[Array[Long]] && fillFactor > 1) {
|
||||
val longBound = bound.asInstanceOf[Array[Long]]
|
||||
for (i <- 0 to bound.length - 1) {
|
||||
for (j <- 0 to fillFactor - 1) {
|
||||
// sample factor shoud not be too large, so it's ok to use 1 / fillfactor as slice
|
||||
newBound(j + i*(fillFactor)) = longBound(i) + (j + 1) * (1 / fillFactor.toDouble)
|
||||
}
|
||||
}
|
||||
(newBound, fillFactor)
|
||||
} else {
|
||||
(bound, 0)
|
||||
}
|
||||
}
|
||||
|
||||
val boundBroadCast = spark.sparkContext.broadcast(expandSampleBoundsWithFactor)
|
||||
|
||||
val indexRdd = rawRdd.mapPartitions { iter =>
|
||||
val expandBoundsWithFactor = boundBroadCast.value
|
||||
val maxBoundNum = expandBoundsWithFactor.map(_._1.length).max
|
||||
val longDecisionBound = new RawDecisionBound(Ordering[Long])
|
||||
val doubleDecisionBound = new RawDecisionBound(Ordering[Double])
|
||||
val stringDecisionBound = new RawDecisionBound(Ordering[String])
|
||||
import java.util.concurrent.ThreadLocalRandom
|
||||
val threadLocalRandom = ThreadLocalRandom.current
|
||||
|
||||
def getRank(rawIndex: Int, value: Long, isNull: Boolean): Int = {
|
||||
val (expandBound, factor) = expandBoundsWithFactor(rawIndex)
|
||||
if (isNull) {
|
||||
expandBound.length + 1
|
||||
} else {
|
||||
if (factor > 1) {
|
||||
doubleDecisionBound.getBound(value + (threadLocalRandom.nextInt(factor) + 1)*(1 / factor.toDouble), expandBound.asInstanceOf[Array[Double]])
|
||||
} else {
|
||||
longDecisionBound.getBound(value, expandBound.asInstanceOf[Array[Long]])
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
iter.map { row =>
|
||||
val values = zFields.zipWithIndex.map { case ((index, field), rawIndex) =>
|
||||
field.dataType match {
|
||||
case LongType =>
|
||||
val isNull = row.isNullAt(index)
|
||||
getRank(rawIndex, if (isNull) 0 else row.getLong(index), isNull)
|
||||
case DoubleType =>
|
||||
val isNull = row.isNullAt(index)
|
||||
getRank(rawIndex, if (isNull) 0 else java.lang.Double.doubleToLongBits(row.getDouble(index)), isNull)
|
||||
case IntegerType =>
|
||||
val isNull = row.isNullAt(index)
|
||||
getRank(rawIndex, if (isNull) 0 else row.getInt(index).toLong, isNull)
|
||||
case FloatType =>
|
||||
val isNull = row.isNullAt(index)
|
||||
getRank(rawIndex, if (isNull) 0 else java.lang.Double.doubleToLongBits(row.getFloat(index).toDouble), isNull)
|
||||
case StringType =>
|
||||
val factor = maxBoundNum.toDouble / expandBoundsWithFactor(rawIndex)._1.length
|
||||
if (row.isNullAt(index)) {
|
||||
maxBoundNum + 1
|
||||
} else {
|
||||
val currentRank = stringDecisionBound.getBound(row.getString(index), expandBoundsWithFactor(rawIndex)._1.asInstanceOf[Array[String]])
|
||||
if (factor > 1) {
|
||||
(currentRank*factor).toInt + threadLocalRandom.nextInt(factor.toInt)
|
||||
} else {
|
||||
currentRank
|
||||
}
|
||||
}
|
||||
case DateType =>
|
||||
val isNull = row.isNullAt(index)
|
||||
getRank(rawIndex, if (isNull) 0 else row.getDate(index).getTime, isNull)
|
||||
case TimestampType =>
|
||||
val isNull = row.isNullAt(index)
|
||||
getRank(rawIndex, if (isNull) 0 else row.getTimestamp(index).getTime, isNull)
|
||||
case ByteType =>
|
||||
val isNull = row.isNullAt(index)
|
||||
getRank(rawIndex, if (isNull) 0 else row.getByte(index).toLong, isNull)
|
||||
case ShortType =>
|
||||
val isNull = row.isNullAt(index)
|
||||
getRank(rawIndex, if (isNull) 0 else row.getShort(index).toLong, isNull)
|
||||
case d: DecimalType =>
|
||||
val isNull = row.isNullAt(index)
|
||||
getRank(rawIndex, if (isNull) 0 else row.getDecimal(index).longValue(), isNull)
|
||||
case _ =>
|
||||
-1
|
||||
}
|
||||
}.filter(v => v != -1).map(ZOrderingUtil.intTo8Byte(_)).toArray
|
||||
val zValues = ZOrderingUtil.interleaving(values, 8)
|
||||
Row.fromSeq(row.toSeq ++ Seq(zValues))
|
||||
}
|
||||
}.sortBy(x => ZorderingBinarySort(x.getAs[Array[Byte]](fieldNum)), numPartitions = fileNum)
|
||||
val newDF = df.sparkSession.createDataFrame(indexRdd, StructType(
|
||||
df.schema.fields ++ Seq(
|
||||
StructField(s"zindex",
|
||||
BinaryType, false))
|
||||
))
|
||||
newDF.drop("zindex")
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user