big-data/hadoop
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

MAPREDUCE-1981. Improve getSplits performance by using listLocatedStatus. Contributed by Hairong Kuang and Jason Lowe

Browse Source 
git-svn-id: https://svn.apache.org/repos/asf/hadoop/common/trunk@1507385 13f79535-47bb-0310-9956-ffa450edef68
This commit is contained in:
Jason Darrell Lowe 2013-07-26 18:16:19 +00:00
parent 55dfadec80
commit ec18984252
6 changed files with 271 additions and 90 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
hadoop-mapreduce-project/CHANGES.txt
View File

@ -156,6 +156,9 @@ Release 2.3.0 - UNRELEASED
OPTIMIZATIONS
MAPREDUCE-1981. Improve getSplits performance by using listLocatedStatus
(Hairong Kuang and Jason Lowe via jlowe)
BUG FIXES
MAPREDUCE-5316. job -list-attempt-ids command does not handle illegal
@ -1232,6 +1235,9 @@ Release 0.23.10 - UNRELEASED
OPTIMIZATIONS
MAPREDUCE-1981. Improve getSplits performance by using listLocatedStatus
(Hairong Kuang and Jason Lowe via jlowe)
BUG FIXES
MAPREDUCE-3193. FileInputFormat doesn't read files recursively in the

89
hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapred/FileInputFormat.java
View File

@ -36,8 +36,10 @@
import org.apache.hadoop.fs.BlockLocation;
import org.apache.hadoop.fs.FileStatus;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.LocatedFileStatus;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.fs.PathFilter;
import org.apache.hadoop.fs.RemoteIterator;
import org.apache.hadoop.mapreduce.security.TokenCache;
import org.apache.hadoop.net.NetworkTopology;
import org.apache.hadoop.net.Node;
@ -169,13 +171,17 @@ public static PathFilter getInputPathFilter(JobConf conf) {
protected void addInputPathRecursively(List<FileStatus> result,
FileSystem fs, Path path, PathFilter inputFilter)
throws IOException {
for(FileStatus stat: fs.listStatus(path, inputFilter)) {
if (stat.isDirectory()) {
addInputPathRecursively(result, fs, stat.getPath(), inputFilter);
} else {
result.add(stat);
RemoteIterator<LocatedFileStatus> iter = fs.listLocatedStatus(path);
while (iter.hasNext()) {
LocatedFileStatus stat = iter.next();
if (inputFilter.accept(stat.getPath())) {
if (stat.isDirectory()) {
addInputPathRecursively(result, fs, stat.getPath(), inputFilter);
} else {
result.add(stat);
}
}
}
}
}
/** List input directories.
@ -221,14 +227,19 @@ protected FileStatus[] listStatus(JobConf job) throws IOException {
} else {
for (FileStatus globStat: matches) {
if (globStat.isDirectory()) {
for(FileStatus stat: fs.listStatus(globStat.getPath(),
inputFilter)) {
if (recursive && stat.isDirectory()) {
addInputPathRecursively(result, fs, stat.getPath(), inputFilter);
} else {
result.add(stat);
RemoteIterator<LocatedFileStatus> iter =
fs.listLocatedStatus(globStat.getPath());
while (iter.hasNext()) {
LocatedFileStatus stat = iter.next();
if (inputFilter.accept(stat.getPath())) {
if (recursive && stat.isDirectory()) {
addInputPathRecursively(result, fs, stat.getPath(),
inputFilter);
} else {
result.add(stat);
}
}
}
}
} else {
result.add(globStat);
}
@ -254,7 +265,6 @@ protected FileSplit makeSplit(Path file, long start, long length,
/** Splits files returned by {@link #listStatus(JobConf)} when
* they're too big.*/
@SuppressWarnings("deprecation")
public InputSplit[] getSplits(JobConf job, int numSplits)
throws IOException {
FileStatus[] files = listStatus(job);
@ -278,31 +288,38 @@ public InputSplit[] getSplits(JobConf job, int numSplits)
NetworkTopology clusterMap = new NetworkTopology();
for (FileStatus file: files) {
Path path = file.getPath();
FileSystem fs = path.getFileSystem(job);
long length = file.getLen();
BlockLocation[] blkLocations = fs.getFileBlockLocations(file, 0, length);
if ((length != 0) && isSplitable(fs, path)) {
long blockSize = file.getBlockSize();
long splitSize = computeSplitSize(goalSize, minSize, blockSize);
if (length != 0) {
FileSystem fs = path.getFileSystem(job);
BlockLocation[] blkLocations;
if (file instanceof LocatedFileStatus) {
blkLocations = ((LocatedFileStatus) file).getBlockLocations();
} else {
blkLocations = fs.getFileBlockLocations(file, 0, length);
}
if (isSplitable(fs, path)) {
long blockSize = file.getBlockSize();
long splitSize = computeSplitSize(goalSize, minSize, blockSize);
long bytesRemaining = length;
while (((double) bytesRemaining)/splitSize > SPLIT_SLOP) {
String[] splitHosts = getSplitHosts(blkLocations,
length-bytesRemaining, splitSize, clusterMap);
splits.add(makeSplit(path, length-bytesRemaining, splitSize,
splitHosts));
bytesRemaining -= splitSize;
long bytesRemaining = length;
while (((double) bytesRemaining)/splitSize > SPLIT_SLOP) {
String[] splitHosts = getSplitHosts(blkLocations,
length-bytesRemaining, splitSize, clusterMap);
splits.add(makeSplit(path, length-bytesRemaining, splitSize,
splitHosts));
bytesRemaining -= splitSize;
}
if (bytesRemaining != 0) {
String[] splitHosts = getSplitHosts(blkLocations, length
- bytesRemaining, bytesRemaining, clusterMap);
splits.add(makeSplit(path, length - bytesRemaining, bytesRemaining,
splitHosts));
}
} else {
String[] splitHosts = getSplitHosts(blkLocations,0,length,clusterMap);
splits.add(makeSplit(path, 0, length, splitHosts));
}
if (bytesRemaining != 0) {
String[] splitHosts = getSplitHosts(blkLocations, length
- bytesRemaining, bytesRemaining, clusterMap);
splits.add(makeSplit(path, length - bytesRemaining, bytesRemaining,
splitHosts));
}
} else if (length != 0) {
String[] splitHosts = getSplitHosts(blkLocations,0,length,clusterMap);
splits.add(makeSplit(path, 0, length, splitHosts));
} else {
//Create empty hosts array for zero length files
splits.add(makeSplit(path, 0, length, new String[0]));

69
hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapreduce/lib/input/CombineFileInputFormat.java
View File

@ -21,7 +21,6 @@
import java.io.IOException;
import java.util.ArrayList;
import java.util.Collection;
import java.util.LinkedList;
import java.util.HashSet;
import java.util.List;
import java.util.HashMap;
@ -33,7 +32,7 @@
import org.apache.hadoop.classification.InterfaceStability;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.FileUtil;
import org.apache.hadoop.fs.LocatedFileStatus;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.fs.BlockLocation;
import org.apache.hadoop.fs.FileStatus;
@ -204,46 +203,33 @@ public List<InputSplit> getSplits(JobContext job)
}
// all the files in input set
Path[] paths = FileUtil.stat2Paths(
listStatus(job).toArray(new FileStatus[0]));
List<FileStatus> stats = listStatus(job);
List<InputSplit> splits = new ArrayList<InputSplit>();
if (paths.length == 0) {
if (stats.size() == 0) {
return splits;
}
// Convert them to Paths first. This is a costly operation and
// we should do it first, otherwise we will incur doing it multiple
// times, one time each for each pool in the next loop.
List<Path> newpaths = new LinkedList<Path>();
for (int i = 0; i < paths.length; i++) {
FileSystem fs = paths[i].getFileSystem(conf);
Path p = fs.makeQualified(paths[i]);
newpaths.add(p);
}
// In one single iteration, process all the paths in a single pool.
// Processing one pool at a time ensures that a split contains paths
// from a single pool only.
for (MultiPathFilter onepool : pools) {
ArrayList<Path> myPaths = new ArrayList<Path>();
ArrayList<FileStatus> myPaths = new ArrayList<FileStatus>();
// pick one input path. If it matches all the filters in a pool,
// add it to the output set
for (Iterator<Path> iter = newpaths.iterator(); iter.hasNext();) {
Path p = iter.next();
if (onepool.accept(p)) {
for (Iterator<FileStatus> iter = stats.iterator(); iter.hasNext();) {
FileStatus p = iter.next();
if (onepool.accept(p.getPath())) {
myPaths.add(p); // add it to my output set
iter.remove();
}
}
// create splits for all files in this pool.
getMoreSplits(job, myPaths.toArray(new Path[myPaths.size()]),
maxSize, minSizeNode, minSizeRack, splits);
getMoreSplits(job, myPaths, maxSize, minSizeNode, minSizeRack, splits);
}
// create splits for all files that are not in any pool.
getMoreSplits(job, newpaths.toArray(new Path[newpaths.size()]),
maxSize, minSizeNode, minSizeRack, splits);
getMoreSplits(job, stats, maxSize, minSizeNode, minSizeRack, splits);
// free up rackToNodes map
rackToNodes.clear();
@ -253,7 +239,7 @@ public List<InputSplit> getSplits(JobContext job)
/**
* Return all the splits in the specified set of paths
*/
private void getMoreSplits(JobContext job, Path[] paths,
private void getMoreSplits(JobContext job, List<FileStatus> stats,
long maxSize, long minSizeNode, long minSizeRack,
List<InputSplit> splits)
throws IOException {
@ -274,15 +260,16 @@ private void getMoreSplits(JobContext job, Path[] paths,
HashMap<String, List<OneBlockInfo>> nodeToBlocks =
new HashMap<String, List<OneBlockInfo>>();
files = new OneFileInfo[paths.length];
if (paths.length == 0) {
files = new OneFileInfo[stats.size()];
if (stats.size() == 0) {
return;
}
// populate all the blocks for all files
long totLength = 0;
for (int i = 0; i < paths.length; i++) {
files[i] = new OneFileInfo(paths[i], conf, isSplitable(job, paths[i]),
int i = 0;
for (FileStatus stat : stats) {
files[i] = new OneFileInfo(stat, conf, isSplitable(job, stat.getPath()),
rackToBlocks, blockToNodes, nodeToBlocks,
rackToNodes, maxSize);
totLength += files[i].getLength();
@ -523,7 +510,7 @@ static class OneFileInfo {
private long fileSize; // size of the file
private OneBlockInfo[] blocks; // all blocks in this file
OneFileInfo(Path path, Configuration conf,
OneFileInfo(FileStatus stat, Configuration conf,
boolean isSplitable,
HashMap<String, List<OneBlockInfo>> rackToBlocks,
HashMap<OneBlockInfo, String[]> blockToNodes,
@ -534,10 +521,13 @@ static class OneFileInfo {
this.fileSize = 0;
// get block locations from file system
FileSystem fs = path.getFileSystem(conf);
FileStatus stat = fs.getFileStatus(path);
BlockLocation[] locations = fs.getFileBlockLocations(stat, 0,
stat.getLen());
BlockLocation[] locations;
if (stat instanceof LocatedFileStatus) {
locations = ((LocatedFileStatus) stat).getBlockLocations();
} else {
FileSystem fs = stat.getPath().getFileSystem(conf);
locations = fs.getFileBlockLocations(stat, 0, stat.getLen());
}
// create a list of all block and their locations
if (locations == null) {
blocks = new OneBlockInfo[0];
@ -552,8 +542,8 @@ static class OneFileInfo {
// full file length
blocks = new OneBlockInfo[1];
fileSize = stat.getLen();
blocks[0] = new OneBlockInfo(path, 0, fileSize, locations[0]
.getHosts(), locations[0].getTopologyPaths());
blocks[0] = new OneBlockInfo(stat.getPath(), 0, fileSize,
locations[0].getHosts(), locations[0].getTopologyPaths());
} else {
ArrayList<OneBlockInfo> blocksList = new ArrayList<OneBlockInfo>(
locations.length);
@ -579,9 +569,9 @@ static class OneFileInfo {
myLength = Math.min(maxSize, left);
}
}
OneBlockInfo oneblock = new OneBlockInfo(path, myOffset,
myLength, locations[i].getHosts(), locations[i]
.getTopologyPaths());
OneBlockInfo oneblock = new OneBlockInfo(stat.getPath(),
myOffset, myLength, locations[i].getHosts(),
locations[i].getTopologyPaths());
left -= myLength;
myOffset += myLength;
@ -693,6 +683,9 @@ static class OneBlockInfo {
protected BlockLocation[] getFileBlockLocations(
FileSystem fs, FileStatus stat) throws IOException {
if (stat instanceof LocatedFileStatus) {
return ((LocatedFileStatus) stat).getBlockLocations();
}
return fs.getFileBlockLocations(stat, 0, stat.getLen());
}

46
hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapreduce/lib/input/FileInputFormat.java
View File

@ -29,9 +29,11 @@
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FileStatus;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.LocatedFileStatus;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.fs.PathFilter;
import org.apache.hadoop.fs.BlockLocation;
import org.apache.hadoop.fs.RemoteIterator;
import org.apache.hadoop.mapreduce.InputFormat;
import org.apache.hadoop.mapreduce.InputSplit;
import org.apache.hadoop.mapreduce.Job;
@ -259,14 +261,19 @@ protected List<FileStatus> listStatus(JobContext job
} else {
for (FileStatus globStat: matches) {
if (globStat.isDirectory()) {
for(FileStatus stat: fs.listStatus(globStat.getPath(),
inputFilter)) {
if (recursive && stat.isDirectory()) {
addInputPathRecursively(result, fs, stat.getPath(), inputFilter);
} else {
result.add(stat);
RemoteIterator<LocatedFileStatus> iter =
fs.listLocatedStatus(globStat.getPath());
while (iter.hasNext()) {
LocatedFileStatus stat = iter.next();
if (inputFilter.accept(stat.getPath())) {
if (recursive && stat.isDirectory()) {
addInputPathRecursively(result, fs, stat.getPath(),
inputFilter);
} else {
result.add(stat);
}
}
}
}
} else {
result.add(globStat);
}
@ -296,13 +303,17 @@ protected List<FileStatus> listStatus(JobContext job
protected void addInputPathRecursively(List<FileStatus> result,
FileSystem fs, Path path, PathFilter inputFilter)
throws IOException {
for(FileStatus stat: fs.listStatus(path, inputFilter)) {
if (stat.isDirectory()) {
addInputPathRecursively(result, fs, stat.getPath(), inputFilter);
} else {
result.add(stat);
RemoteIterator<LocatedFileStatus> iter = fs.listLocatedStatus(path);
while (iter.hasNext()) {
LocatedFileStatus stat = iter.next();
if (inputFilter.accept(stat.getPath())) {
if (stat.isDirectory()) {
addInputPathRecursively(result, fs, stat.getPath(), inputFilter);
} else {
result.add(stat);
}
}
}
}
}
@ -331,8 +342,13 @@ public List<InputSplit> getSplits(JobContext job) throws IOException {
Path path = file.getPath();
long length = file.getLen();
if (length != 0) {
FileSystem fs = path.getFileSystem(job.getConfiguration());
BlockLocation[] blkLocations = fs.getFileBlockLocations(file, 0, length);
BlockLocation[] blkLocations;
if (file instanceof LocatedFileStatus) {
blkLocations = ((LocatedFileStatus) file).getBlockLocations();
} else {
FileSystem fs = path.getFileSystem(job.getConfiguration());
blkLocations = fs.getFileBlockLocations(file, 0, length);
}
if (isSplitable(job, path)) {
long blockSize = file.getBlockSize();
long splitSize = computeSplitSize(blockSize, minSize, maxSize);

113
hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/test/java/org/apache/hadoop/mapred/TestFileInputFormat.java Normal file
View File

@ -0,0 +1,113 @@
/**
* 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.hadoop.mapred;
import java.io.FileNotFoundException;
import java.io.IOException;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.BlockLocation;
import org.apache.hadoop.fs.FileStatus;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.LocatedFileStatus;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.fs.PathFilter;
import org.apache.hadoop.fs.RawLocalFileSystem;
import org.apache.hadoop.fs.RemoteIterator;
import org.junit.Assert;
import org.junit.Test;
public class TestFileInputFormat {
@Test
public void testListLocatedStatus() throws Exception {
Configuration conf = getConfiguration();
conf.setBoolean("fs.test.impl.disable.cache", false);
conf.set(org.apache.hadoop.mapreduce.lib.input.FileInputFormat.INPUT_DIR,
"test:///a1/a2");
MockFileSystem mockFs =
(MockFileSystem) new Path("test:///").getFileSystem(conf);
Assert.assertEquals("listLocatedStatus already called",
0, mockFs.numListLocatedStatusCalls);
JobConf job = new JobConf(conf);
TextInputFormat fileInputFormat = new TextInputFormat();
fileInputFormat.configure(job);
InputSplit[] splits = fileInputFormat.getSplits(job, 1);
Assert.assertEquals("Input splits are not correct", 2, splits.length);
Assert.assertEquals("listLocatedStatuss calls",
1, mockFs.numListLocatedStatusCalls);
}
private Configuration getConfiguration() {
Configuration conf = new Configuration();
conf.set("fs.test.impl.disable.cache", "true");
conf.setClass("fs.test.impl", MockFileSystem.class, FileSystem.class);
conf.set(org.apache.hadoop.mapreduce.lib.input.FileInputFormat.INPUT_DIR,
"test:///a1");
return conf;
}
static class MockFileSystem extends RawLocalFileSystem {
int numListLocatedStatusCalls = 0;
@Override
public FileStatus[] listStatus(Path f) throws FileNotFoundException,
IOException {
if (f.toString().equals("test:/a1")) {
return new FileStatus[] {
new FileStatus(0, true, 1, 150, 150, new Path("test:/a1/a2")),
new FileStatus(10, false, 1, 150, 150, new Path("test:/a1/file1")) };
} else if (f.toString().equals("test:/a1/a2")) {
return new FileStatus[] {
new FileStatus(10, false, 1, 150, 150,
new Path("test:/a1/a2/file2")),
new FileStatus(10, false, 1, 151, 150,
new Path("test:/a1/a2/file3")) };
}
return new FileStatus[0];
}
@Override
public FileStatus[] globStatus(Path pathPattern, PathFilter filter)
throws IOException {
return new FileStatus[] { new FileStatus(10, true, 1, 150, 150,
pathPattern) };
}
@Override
public FileStatus[] listStatus(Path f, PathFilter filter)
throws FileNotFoundException, IOException {
return this.listStatus(f);
}
@Override
public BlockLocation[] getFileBlockLocations(Path p, long start, long len)
throws IOException {
return new BlockLocation[] {
new BlockLocation(new String[] { "localhost:50010" },
new String[] { "localhost" }, 0, len) };
}
@Override
protected RemoteIterator<LocatedFileStatus> listLocatedStatus(Path f,
PathFilter filter) throws FileNotFoundException, IOException {
++numListLocatedStatusCalls;
return super.listLocatedStatus(f, filter);
}
}
}

38
hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/test/java/org/apache/hadoop/mapreduce/lib/input/TestFileInputFormat.java
View File

@ -24,11 +24,14 @@
import junit.framework.Assert;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.BlockLocation;
import org.apache.hadoop.fs.FileStatus;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.LocatedFileStatus;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.fs.PathFilter;
import org.apache.hadoop.fs.RawLocalFileSystem;
import org.apache.hadoop.fs.RemoteIterator;
import org.apache.hadoop.mapreduce.InputSplit;
import org.apache.hadoop.mapreduce.Job;
import org.junit.Test;
@ -77,6 +80,23 @@ public void testNumInputFilesWithoutRecursively() throws Exception {
.toString());
}
@Test
public void testListLocatedStatus() throws Exception {
Configuration conf = getConfiguration();
conf.setBoolean("fs.test.impl.disable.cache", false);
conf.set(FileInputFormat.INPUT_DIR, "test:///a1/a2");
MockFileSystem mockFs =
(MockFileSystem) new Path("test:///").getFileSystem(conf);
Assert.assertEquals("listLocatedStatus already called",
0, mockFs.numListLocatedStatusCalls);
Job job = Job.getInstance(conf);
FileInputFormat<?, ?> fileInputFormat = new TextInputFormat();
List<InputSplit> splits = fileInputFormat.getSplits(job);
Assert.assertEquals("Input splits are not correct", 2, splits.size());
Assert.assertEquals("listLocatedStatuss calls",
1, mockFs.numListLocatedStatusCalls);
}
private Configuration getConfiguration() {
Configuration conf = new Configuration();
conf.set("fs.test.impl.disable.cache", "true");
@ -86,13 +106,14 @@ private Configuration getConfiguration() {
}
static class MockFileSystem extends RawLocalFileSystem {
int numListLocatedStatusCalls = 0;
@Override
public FileStatus[] listStatus(Path f) throws FileNotFoundException,
IOException {
if (f.toString().equals("test:/a1")) {
return new FileStatus[] {
new FileStatus(10, true, 1, 150, 150, new Path("test:/a1/a2")),
new FileStatus(0, true, 1, 150, 150, new Path("test:/a1/a2")),
new FileStatus(10, false, 1, 150, 150, new Path("test:/a1/file1")) };
} else if (f.toString().equals("test:/a1/a2")) {
return new FileStatus[] {
@ -116,5 +137,20 @@ public FileStatus[] listStatus(Path f, PathFilter filter)
throws FileNotFoundException, IOException {
return this.listStatus(f);
}
@Override
public BlockLocation[] getFileBlockLocations(Path p, long start, long len)
throws IOException {
return new BlockLocation[] {
new BlockLocation(new String[] { "localhost:50010" },
new String[] { "localhost" }, 0, len) };
}
@Override
protected RemoteIterator<LocatedFileStatus> listLocatedStatus(Path f,
PathFilter filter) throws FileNotFoundException, IOException {
++numListLocatedStatusCalls;
return super.listLocatedStatus(f, filter);
}
}
}