MAPREDUCE-5478. TeraInputFormat unnecessarily defines its own FileSplit subclass (Sandy Ryza)

git-svn-id: https://svn.apache.org/repos/asf/hadoop/common/trunk@1517046 13f79535-47bb-0310-9956-ffa450edef68
This commit is contained in:
Sanford Ryza 2013-08-23 21:23:10 +00:00
parent 15632cd76f
commit 9ee38f3a84
3 changed files with 7 additions and 51 deletions

View File

@ -181,6 +181,9 @@ Release 2.1.1-beta - UNRELEASED
IMPROVEMENTS
MAPREDUCE-5478. TeraInputFormat unnecessarily defines its own FileSplit
subclass (Sandy Ryza)
OPTIMIZATIONS
MAPREDUCE-5446. TestJobHistoryEvents and TestJobHistoryParsing have race

View File

@ -60,48 +60,6 @@ public class TeraInputFormat extends FileInputFormat<Text,Text> {
private static MRJobConfig lastContext = null;
private static List<InputSplit> lastResult = null;
static class TeraFileSplit extends FileSplit {
static private String[] ZERO_LOCATIONS = new String[0];
private String[] locations;
public TeraFileSplit() {
locations = ZERO_LOCATIONS;
}
public TeraFileSplit(Path file, long start, long length, String[] hosts) {
super(file, start, length, hosts);
try {
locations = super.getLocations();
} catch (IOException e) {
locations = ZERO_LOCATIONS;
}
}
// XXXXXX should this also be null-protected?
protected void setLocations(String[] hosts) {
locations = hosts;
}
@Override
public String[] getLocations() {
return locations;
}
public String toString() {
StringBuffer result = new StringBuffer();
result.append(getPath());
result.append(" from ");
result.append(getStart());
result.append(" length ");
result.append(getLength());
for(String host: getLocations()) {
result.append(" ");
result.append(host);
}
return result.toString();
}
}
static class TextSampler implements IndexedSortable {
private ArrayList<Text> records = new ArrayList<Text>();
@ -325,11 +283,6 @@ public boolean nextKeyValue() throws IOException {
return new TeraRecordReader();
}
protected FileSplit makeSplit(Path file, long start, long length,
String[] hosts) {
return new TeraFileSplit(file, start, length, hosts);
}
@Override
public List<InputSplit> getSplits(JobContext job) throws IOException {
if (job == lastContext) {
@ -343,7 +296,7 @@ public List<InputSplit> getSplits(JobContext job) throws IOException {
System.out.println("Spent " + (t2 - t1) + "ms computing base-splits.");
if (job.getConfiguration().getBoolean(TeraScheduler.USE, true)) {
TeraScheduler scheduler = new TeraScheduler(
lastResult.toArray(new TeraFileSplit[0]), job.getConfiguration());
lastResult.toArray(new FileSplit[0]), job.getConfiguration());
lastResult = scheduler.getNewFileSplits();
t3 = System.currentTimeMillis();
System.out.println("Spent " + (t3 - t2) + "ms computing TeraScheduler splits.");

View File

@ -24,7 +24,6 @@
import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.examples.terasort.TeraInputFormat.TeraFileSplit;
import org.apache.hadoop.mapreduce.InputSplit;
import org.apache.hadoop.mapreduce.lib.input.FileSplit;
import org.apache.hadoop.mapreduce.server.tasktracker.TTConfig;
@ -214,8 +213,9 @@ public List<InputSplit> getNewFileSplits() throws IOException {
for(int i=0; i < splits.length; ++i) {
if (splits[i].isAssigned) {
// copy the split and fix up the locations
((TeraFileSplit) realSplits[i]).setLocations
(new String[]{splits[i].locations.get(0).hostname});
String[] newLocations = {splits[i].locations.get(0).hostname};
realSplits[i] = new FileSplit(realSplits[i].getPath(),
realSplits[i].getStart(), realSplits[i].getLength(), newLocations);
result[left++] = realSplits[i];
} else {
result[right--] = realSplits[i];