org.apache.hadoop.hbase.mapreduce

Class TableSnapshotInputFormatImpl

java.lang.Object

org.apache.hadoop.hbase.mapreduce.TableSnapshotInputFormatImpl

@InterfaceAudience.Private
public class TableSnapshotInputFormatImpl
extends Object

Hadoop MR API-agnostic implementation for mapreduce over table snapshots.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	TableSnapshotInputFormatImpl.InputSplit Implementation class for InputSplit logic common between mapred and mapreduce.
static class	TableSnapshotInputFormatImpl.RecordReader Implementation class for RecordReader logic common between mapred and mapreduce.

Field Summary

Fields

Modifier and Type	Field and Description
private static float	DEFAULT_LOCALITY_CUTOFF_MULTIPLIER
private static String	LOCALITY_CUTOFF_MULTIPLIER See getBestLocations(Configuration, HDFSBlocksDistribution, int)
static org.slf4j.Logger	LOG
static String	NUM_SPLITS_PER_REGION For MapReduce jobs running multiple mappers per region, determines number of splits to generate per region.
protected static String	RESTORE_DIR_KEY
static boolean	SNAPSHOT_INPUTFORMAT_LOCALITY_ENABLED_DEFAULT
static String	SNAPSHOT_INPUTFORMAT_LOCALITY_ENABLED_KEY Whether to calculate the block location for splits.
private static String	SNAPSHOT_NAME_KEY
static String	SPLIT_ALGO For MapReduce jobs running multiple mappers per region, determines what split algorithm we should be using to find split points for scanners.

Constructor Summary

Constructors

Constructor and Description
TableSnapshotInputFormatImpl()

Method Summary

Modifier and Type	Method and Description
private static List<String>	calculateLocationsForInputSplit(org.apache.hadoop.conf.Configuration conf, TableDescriptor htd, HRegionInfo hri, org.apache.hadoop.fs.Path tableDir, boolean localityEnabled) Compute block locations for snapshot files (which will get the locations for referred hfiles) only when localityEnabled is true.
static Scan	extractScanFromConf(org.apache.hadoop.conf.Configuration conf)
static List<String>	getBestLocations(org.apache.hadoop.conf.Configuration conf, HDFSBlocksDistribution blockDistribution)
private static List<String>	getBestLocations(org.apache.hadoop.conf.Configuration conf, HDFSBlocksDistribution blockDistribution, int numTopsAtMost) This computes the locations to be passed from the InputSplit.
static List<HRegionInfo>	getRegionInfosFromManifest(SnapshotManifest manifest)
static SnapshotManifest	getSnapshotManifest(org.apache.hadoop.conf.Configuration conf, String snapshotName, org.apache.hadoop.fs.Path rootDir, org.apache.hadoop.fs.FileSystem fs)
private static String	getSnapshotName(org.apache.hadoop.conf.Configuration conf)
static RegionSplitter.SplitAlgorithm	getSplitAlgo(org.apache.hadoop.conf.Configuration conf)
static List<TableSnapshotInputFormatImpl.InputSplit>	getSplits(org.apache.hadoop.conf.Configuration conf)
static List<TableSnapshotInputFormatImpl.InputSplit>	getSplits(Scan scan, SnapshotManifest manifest, List<HRegionInfo> regionManifests, org.apache.hadoop.fs.Path restoreDir, org.apache.hadoop.conf.Configuration conf)
static List<TableSnapshotInputFormatImpl.InputSplit>	getSplits(Scan scan, SnapshotManifest manifest, List<HRegionInfo> regionManifests, org.apache.hadoop.fs.Path restoreDir, org.apache.hadoop.conf.Configuration conf, RegionSplitter.SplitAlgorithm sa, int numSplits)
static void	setInput(org.apache.hadoop.conf.Configuration conf, String snapshotName, org.apache.hadoop.fs.Path restoreDir) Configures the job to use TableSnapshotInputFormat to read from a snapshot.
static void	setInput(org.apache.hadoop.conf.Configuration conf, String snapshotName, org.apache.hadoop.fs.Path restoreDir, RegionSplitter.SplitAlgorithm splitAlgo, int numSplitsPerRegion) Configures the job to use TableSnapshotInputFormat to read from a snapshot.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

LOG

public static final org.slf4j.Logger LOG

SNAPSHOT_NAME_KEY

private static final String SNAPSHOT_NAME_KEY

See Also:

Constant Field Values

RESTORE_DIR_KEY

protected static final String RESTORE_DIR_KEY

See Also:

Constant Field Values

LOCALITY_CUTOFF_MULTIPLIER

private static final String LOCALITY_CUTOFF_MULTIPLIER

See getBestLocations(Configuration, HDFSBlocksDistribution, int)

See Also:

Constant Field Values

DEFAULT_LOCALITY_CUTOFF_MULTIPLIER

private static final float DEFAULT_LOCALITY_CUTOFF_MULTIPLIER

See Also:

Constant Field Values

SPLIT_ALGO

public static final String SPLIT_ALGO

For MapReduce jobs running multiple mappers per region, determines what split algorithm we should be using to find split points for scanners.

See Also:

Constant Field Values

NUM_SPLITS_PER_REGION

public static final String NUM_SPLITS_PER_REGION

For MapReduce jobs running multiple mappers per region, determines number of splits to generate per region.

See Also:

Constant Field Values

SNAPSHOT_INPUTFORMAT_LOCALITY_ENABLED_KEY

public static final String SNAPSHOT_INPUTFORMAT_LOCALITY_ENABLED_KEY

Whether to calculate the block location for splits. Default to true. If the computing layer runs outside of HBase cluster, the block locality does not master. Setting this value to false could skip the calculation and save some time. Set access modifier to "public" so that these could be accessed by test classes of both org.apache.hadoop.hbase.mapred and org.apache.hadoop.hbase.mapreduce.

See Also:

Constant Field Values

SNAPSHOT_INPUTFORMAT_LOCALITY_ENABLED_DEFAULT

public static final boolean SNAPSHOT_INPUTFORMAT_LOCALITY_ENABLED_DEFAULT

See Also:

Constant Field Values

Constructor Detail

TableSnapshotInputFormatImpl

public TableSnapshotInputFormatImpl()

Method Detail

getSplits

public static List<TableSnapshotInputFormatImpl.InputSplit> getSplits(org.apache.hadoop.conf.Configuration conf)
                                                               throws IOException

Throws:

IOException

getSplitAlgo

public static RegionSplitter.SplitAlgorithm getSplitAlgo(org.apache.hadoop.conf.Configuration conf)
                                                  throws IOException

Throws:

IOException

getRegionInfosFromManifest

public static List<HRegionInfo> getRegionInfosFromManifest(SnapshotManifest manifest)

getSnapshotManifest

public static SnapshotManifest getSnapshotManifest(org.apache.hadoop.conf.Configuration conf,
                                                   String snapshotName,
                                                   org.apache.hadoop.fs.Path rootDir,
                                                   org.apache.hadoop.fs.FileSystem fs)
                                            throws IOException

Throws:

IOException

extractScanFromConf

public static Scan extractScanFromConf(org.apache.hadoop.conf.Configuration conf)
                                throws IOException

Throws:

IOException

getSplits

public static List<TableSnapshotInputFormatImpl.InputSplit> getSplits(Scan scan,
                                                                      SnapshotManifest manifest,
                                                                      List<HRegionInfo> regionManifests,
                                                                      org.apache.hadoop.fs.Path restoreDir,
                                                                      org.apache.hadoop.conf.Configuration conf)
                                                               throws IOException

Throws:

IOException

getSplits

public static List<TableSnapshotInputFormatImpl.InputSplit> getSplits(Scan scan,
                                                                      SnapshotManifest manifest,
                                                                      List<HRegionInfo> regionManifests,
                                                                      org.apache.hadoop.fs.Path restoreDir,
                                                                      org.apache.hadoop.conf.Configuration conf,
                                                                      RegionSplitter.SplitAlgorithm sa,
                                                                      int numSplits)
                                                               throws IOException

Throws:

IOException

calculateLocationsForInputSplit

private static List<String> calculateLocationsForInputSplit(org.apache.hadoop.conf.Configuration conf,
                                                            TableDescriptor htd,
                                                            HRegionInfo hri,
                                                            org.apache.hadoop.fs.Path tableDir,
                                                            boolean localityEnabled)
                                                     throws IOException

Compute block locations for snapshot files (which will get the locations for referred hfiles) only when localityEnabled is true.

Throws:

IOException

getBestLocations

private static List<String> getBestLocations(org.apache.hadoop.conf.Configuration conf,
                                             HDFSBlocksDistribution blockDistribution,
                                             int numTopsAtMost)

This computes the locations to be passed from the InputSplit. MR/Yarn schedulers does not take weights into account, thus will treat every location passed from the input split as equal. We do not want to blindly pass all the locations, since we are creating one split per region, and the region's blocks are all distributed throughout the cluster unless favorite node assignment is used. On the expected stable case, only one location will contain most of the blocks as local. On the other hand, in favored node assignment, 3 nodes will contain highly local blocks. Here we are doing a simple heuristic, where we will pass all hosts which have at least 80% (hbase.tablesnapshotinputformat.locality.cutoff.multiplier) as much block locality as the top host with the best locality. Return at most numTopsAtMost locations if there are more than that.

getBestLocations

public static List<String> getBestLocations(org.apache.hadoop.conf.Configuration conf,
                                            HDFSBlocksDistribution blockDistribution)

getSnapshotName

private static String getSnapshotName(org.apache.hadoop.conf.Configuration conf)

setInput

public static void setInput(org.apache.hadoop.conf.Configuration conf,
                            String snapshotName,
                            org.apache.hadoop.fs.Path restoreDir)
                     throws IOException

Configures the job to use TableSnapshotInputFormat to read from a snapshot.

Parameters:

conf - the job to configuration

snapshotName - the name of the snapshot to read from

restoreDir - a temporary directory to restore the snapshot into. Current user should have write permissions to this directory, and this should not be a subdirectory of rootdir. After the job is finished, restoreDir can be deleted.

Throws:

IOException - if an error occurs

setInput

public static void setInput(org.apache.hadoop.conf.Configuration conf,
                            String snapshotName,
                            org.apache.hadoop.fs.Path restoreDir,
                            RegionSplitter.SplitAlgorithm splitAlgo,
                            int numSplitsPerRegion)
                     throws IOException

Configures the job to use TableSnapshotInputFormat to read from a snapshot.

Parameters:

conf - the job to configure

snapshotName - the name of the snapshot to read from

restoreDir - a temporary directory to restore the snapshot into. Current user should have write permissions to this directory, and this should not be a subdirectory of rootdir. After the job is finished, restoreDir can be deleted.

numSplitsPerRegion - how many input splits to generate per one region

splitAlgo - SplitAlgorithm to be used when generating InputSplits

Throws:

IOException - if an error occurs