[SPARK-23817][SQL] Create file source V2 framework and migrate ORC read path #23383
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -373,8 +373,7 @@ case class FileSourceScanExec( | |
| val filesGroupedToBuckets = | ||
| selectedPartitions.flatMap { p => | ||
| p.files.filter(_.getLen > 0).map { f => | ||
| PartitionedFileUtil.getPartitionedFile(f, f.getPath, p.values) | ||
| } | ||
| }.groupBy { f => | ||
| BucketingUtils | ||
|
|
@@ -410,107 +409,35 @@ case class FileSourceScanExec( | |
| readFile: (PartitionedFile) => Iterator[InternalRow], | ||
| selectedPartitions: Seq[PartitionDirectory], | ||
| fsRelation: HadoopFsRelation): RDD[InternalRow] = { | ||
| val openCostInBytes = fsRelation.sparkSession.sessionState.conf.filesOpenCostInBytes | ||
| val maxSplitBytes = | ||
| FilePartition.maxSplitBytes(fsRelation.sparkSession, selectedPartitions) | ||
| logInfo(s"Planning scan with bin packing, max size: $maxSplitBytes bytes, " + | ||
| s"open cost is considered as scanning $openCostInBytes bytes.") | ||
|
|
||
| val splitFiles = selectedPartitions.flatMap { partition => | ||
| partition.files.filter(_.getLen > 0).flatMap { file => | ||
| // getPath() is very expensive so we only want to call it once in this block: | ||
| val filePath = file.getPath | ||
|
Member
There was a problem hiding this comment. @gengliangwang I'd like to understand why "getPath() is very expensive", it seems just a simple field access, should be relatively cheap.
Member
Author
There was a problem hiding this comment. It can be expensive on object storage services like aws s3
Member
There was a problem hiding this comment. @gengliangwang Could you please elaborate more? I didn't find anything special for I noticed this comment when working on SPARK-52011 (#50765). The change brings many benefits for HDFS, but I don't have an object storage environment to test. |
||
| val isSplitable = relation.fileFormat.isSplitable( | ||
| relation.sparkSession, relation.options, filePath) | ||
| PartitionedFileUtil.splitFiles( | ||
| sparkSession = relation.sparkSession, | ||
| file = file, | ||
| filePath = filePath, | ||
| isSplitable = isSplitable, | ||
| maxSplitBytes = maxSplitBytes, | ||
| partitionValues = partition.values | ||
| ) | ||
| } | ||
| }.toArray.sortBy(_.length)(implicitly[Ordering[Long]].reverse) | ||
|
|
||
| val partitions = | ||
| FilePartition.getFilePartitions(relation.sparkSession, splitFiles, maxSplitBytes) | ||
|
|
||
| new FileScanRDD(fsRelation.sparkSession, readFile, partitions) | ||
| } | ||
|
|
||
| override def doCanonicalize(): FileSourceScanExec = { | ||
| FileSourceScanExec( | ||
| relation, | ||
|
|
||
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,94 @@ | ||
| /* | ||
| * 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.execution | ||
|
|
||
| import org.apache.hadoop.fs.{BlockLocation, FileStatus, LocatedFileStatus, Path} | ||
|
|
||
| import org.apache.spark.sql.SparkSession | ||
| import org.apache.spark.sql.catalyst.InternalRow | ||
| import org.apache.spark.sql.execution.datasources._ | ||
|
|
||
| object PartitionedFileUtil { | ||
|
||
| def splitFiles( | ||
| sparkSession: SparkSession, | ||
| file: FileStatus, | ||
| filePath: Path, | ||
| isSplitable: Boolean, | ||
| maxSplitBytes: Long, | ||
| partitionValues: InternalRow): Seq[PartitionedFile] = { | ||
| if (isSplitable) { | ||
| (0L until file.getLen by maxSplitBytes).map { offset => | ||
| val remaining = file.getLen - offset | ||
| val size = if (remaining > maxSplitBytes) maxSplitBytes else remaining | ||
| val hosts = getBlockHosts(getBlockLocations(file), offset, size) | ||
| PartitionedFile(partitionValues, filePath.toUri.toString, offset, size, hosts) | ||
| } | ||
| } else { | ||
| Seq(getPartitionedFile(file, filePath, partitionValues)) | ||
| } | ||
| } | ||
|
|
||
| def getPartitionedFile( | ||
| file: FileStatus, | ||
| filePath: Path, | ||
| partitionValues: InternalRow): PartitionedFile = { | ||
| val hosts = getBlockHosts(getBlockLocations(file), 0, file.getLen) | ||
| PartitionedFile(partitionValues, filePath.toUri.toString, 0, file.getLen, hosts) | ||
| } | ||
|
|
||
| private def getBlockLocations(file: FileStatus): Array[BlockLocation] = file match { | ||
| case f: LocatedFileStatus => f.getBlockLocations | ||
| case f => Array.empty[BlockLocation] | ||
| } | ||
| // Given locations of all blocks of a single file, `blockLocations`, and an `(offset, length)` | ||
| // pair that represents a segment of the same file, find out the block that contains the largest | ||
| // fraction the segment, and returns location hosts of that block. If no such block can be found, | ||
| // returns an empty array. | ||
| private def getBlockHosts( | ||
gengliangwang marked this conversation as resolved.
Outdated
Show resolved
Hide resolved
|
||
| blockLocations: Array[BlockLocation], | ||
| offset: Long, | ||
| length: Long): Array[String] = { | ||
| val candidates = blockLocations.map { | ||
| // The fragment starts from a position within this block | ||
| case b if b.getOffset <= offset && offset < b.getOffset + b.getLength => | ||
| b.getHosts -> (b.getOffset + b.getLength - offset).min(length) | ||
|
|
||
| // The fragment ends at a position within this block | ||
| case b if offset <= b.getOffset && offset + length < b.getLength => | ||
| b.getHosts -> (offset + length - b.getOffset).min(length) | ||
|
|
||
| // The fragment fully contains this block | ||
| case b if offset <= b.getOffset && b.getOffset + b.getLength <= offset + length => | ||
| b.getHosts -> b.getLength | ||
|
|
||
| // The fragment doesn't intersect with this block | ||
| case b => | ||
| b.getHosts -> 0L | ||
| }.filter { case (hosts, size) => | ||
| size > 0L | ||
| } | ||
|
|
||
| if (candidates.isEmpty) { | ||
| Array.empty[String] | ||
| } else { | ||
| val (hosts, _) = candidates.maxBy { case (_, size) => size } | ||
| hosts | ||
| } | ||
| } | ||
| } | ||
|
|
||
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Given the overall DSv2 risk, shall we make this
publicand add migration doc officially, @gatorsmile ?There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
the streaming flag is also internal, if we want to change it, let's change them together in a followup PR.