HBASE-20226: Parallelize region manifest deletes (#2159)
We observed this delete call to be a bottleneck for table with lots of
regions. Patch attempts to parallelize them.
Signed-off-by: Andrew Purtell <apurtell@apache.org>
(cherry picked from commit f07f30ae24
)
This commit is contained in:
parent
bba70f08ea
commit
06236dbfcc
|
@ -20,11 +20,15 @@ package org.apache.hadoop.hbase.snapshot;
|
||||||
|
|
||||||
import java.io.FileNotFoundException;
|
import java.io.FileNotFoundException;
|
||||||
import java.io.IOException;
|
import java.io.IOException;
|
||||||
|
import java.io.InterruptedIOException;
|
||||||
import java.util.ArrayList;
|
import java.util.ArrayList;
|
||||||
import java.util.Collection;
|
import java.util.Collection;
|
||||||
import java.util.HashMap;
|
import java.util.HashMap;
|
||||||
import java.util.List;
|
import java.util.List;
|
||||||
import java.util.Map;
|
import java.util.Map;
|
||||||
|
import java.util.concurrent.Callable;
|
||||||
|
import java.util.concurrent.ExecutionException;
|
||||||
|
import java.util.concurrent.ExecutorCompletionService;
|
||||||
import java.util.concurrent.ThreadPoolExecutor;
|
import java.util.concurrent.ThreadPoolExecutor;
|
||||||
import java.util.concurrent.TimeUnit;
|
import java.util.concurrent.TimeUnit;
|
||||||
import org.apache.hadoop.conf.Configuration;
|
import org.apache.hadoop.conf.Configuration;
|
||||||
|
@ -513,44 +517,64 @@ public final class SnapshotManifest {
|
||||||
workingDir, desc);
|
workingDir, desc);
|
||||||
v2Regions = SnapshotManifestV2.loadRegionManifests(conf, tpool, workingDirFs,
|
v2Regions = SnapshotManifestV2.loadRegionManifests(conf, tpool, workingDirFs,
|
||||||
workingDir, desc, manifestSizeLimit);
|
workingDir, desc, manifestSizeLimit);
|
||||||
|
|
||||||
|
SnapshotDataManifest.Builder dataManifestBuilder = SnapshotDataManifest.newBuilder();
|
||||||
|
dataManifestBuilder.setTableSchema(ProtobufUtil.toTableSchema(htd));
|
||||||
|
|
||||||
|
if (v1Regions != null && v1Regions.size() > 0) {
|
||||||
|
dataManifestBuilder.addAllRegionManifests(v1Regions);
|
||||||
|
}
|
||||||
|
if (v2Regions != null && v2Regions.size() > 0) {
|
||||||
|
dataManifestBuilder.addAllRegionManifests(v2Regions);
|
||||||
|
}
|
||||||
|
|
||||||
|
// Write the v2 Data Manifest.
|
||||||
|
// Once the data-manifest is written, the snapshot can be considered complete.
|
||||||
|
// Currently snapshots are written in a "temporary" directory and later
|
||||||
|
// moved to the "complated" snapshot directory.
|
||||||
|
setStatusMsg("Writing data manifest for " + this.desc.getName());
|
||||||
|
SnapshotDataManifest dataManifest = dataManifestBuilder.build();
|
||||||
|
writeDataManifest(dataManifest);
|
||||||
|
this.regionManifests = dataManifest.getRegionManifestsList();
|
||||||
|
|
||||||
|
// Remove the region manifests. Everything is now in the data-manifest.
|
||||||
|
// The delete operation is "relaxed", unless we get an exception we keep going.
|
||||||
|
// The extra files in the snapshot directory will not give any problem,
|
||||||
|
// since they have the same content as the data manifest, and even by re-reading
|
||||||
|
// them we will get the same information.
|
||||||
|
int totalDeletes = 0;
|
||||||
|
ExecutorCompletionService<Void> completionService = new ExecutorCompletionService<>(tpool);
|
||||||
|
if (v1Regions != null) {
|
||||||
|
for (SnapshotRegionManifest regionManifest: v1Regions) {
|
||||||
|
++totalDeletes;
|
||||||
|
completionService.submit(() -> {
|
||||||
|
SnapshotManifestV1.deleteRegionManifest(workingDirFs, workingDir, regionManifest);
|
||||||
|
return null;
|
||||||
|
});
|
||||||
|
}
|
||||||
|
}
|
||||||
|
if (v2Regions != null) {
|
||||||
|
for (SnapshotRegionManifest regionManifest: v2Regions) {
|
||||||
|
++totalDeletes;
|
||||||
|
completionService.submit(() -> {
|
||||||
|
SnapshotManifestV2.deleteRegionManifest(workingDirFs, workingDir, regionManifest);
|
||||||
|
return null;
|
||||||
|
});
|
||||||
|
}
|
||||||
|
}
|
||||||
|
// Wait for the deletes to finish.
|
||||||
|
for (int i = 0; i < totalDeletes; i++) {
|
||||||
|
try {
|
||||||
|
completionService.take().get();
|
||||||
|
} catch (InterruptedException ie) {
|
||||||
|
throw new InterruptedIOException(ie.getMessage());
|
||||||
|
} catch (ExecutionException e) {
|
||||||
|
throw new IOException("Error deleting region manifests", e.getCause());
|
||||||
|
}
|
||||||
|
}
|
||||||
} finally {
|
} finally {
|
||||||
tpool.shutdown();
|
tpool.shutdown();
|
||||||
}
|
}
|
||||||
|
|
||||||
SnapshotDataManifest.Builder dataManifestBuilder = SnapshotDataManifest.newBuilder();
|
|
||||||
dataManifestBuilder.setTableSchema(ProtobufUtil.toTableSchema(htd));
|
|
||||||
|
|
||||||
if (v1Regions != null && v1Regions.size() > 0) {
|
|
||||||
dataManifestBuilder.addAllRegionManifests(v1Regions);
|
|
||||||
}
|
|
||||||
if (v2Regions != null && v2Regions.size() > 0) {
|
|
||||||
dataManifestBuilder.addAllRegionManifests(v2Regions);
|
|
||||||
}
|
|
||||||
|
|
||||||
// Write the v2 Data Manifest.
|
|
||||||
// Once the data-manifest is written, the snapshot can be considered complete.
|
|
||||||
// Currently snapshots are written in a "temporary" directory and later
|
|
||||||
// moved to the "complated" snapshot directory.
|
|
||||||
setStatusMsg("Writing data manifest for " + this.desc.getName());
|
|
||||||
SnapshotDataManifest dataManifest = dataManifestBuilder.build();
|
|
||||||
writeDataManifest(dataManifest);
|
|
||||||
this.regionManifests = dataManifest.getRegionManifestsList();
|
|
||||||
|
|
||||||
// Remove the region manifests. Everything is now in the data-manifest.
|
|
||||||
// The delete operation is "relaxed", unless we get an exception we keep going.
|
|
||||||
// The extra files in the snapshot directory will not give any problem,
|
|
||||||
// since they have the same content as the data manifest, and even by re-reading
|
|
||||||
// them we will get the same information.
|
|
||||||
if (v1Regions != null && v1Regions.size() > 0) {
|
|
||||||
for (SnapshotRegionManifest regionManifest: v1Regions) {
|
|
||||||
SnapshotManifestV1.deleteRegionManifest(workingDirFs, workingDir, regionManifest);
|
|
||||||
}
|
|
||||||
}
|
|
||||||
if (v2Regions != null && v2Regions.size() > 0) {
|
|
||||||
for (SnapshotRegionManifest regionManifest: v2Regions) {
|
|
||||||
SnapshotManifestV2.deleteRegionManifest(workingDirFs, workingDir, regionManifest);
|
|
||||||
}
|
|
||||||
}
|
|
||||||
}
|
}
|
||||||
|
|
||||||
/*
|
/*
|
||||||
|
|
Loading…
Reference in New Issue