org.apache.druid.segment.realtime.appenderator

Class BatchAppenderator

java.lang.Object

org.apache.druid.segment.realtime.appenderator.BatchAppenderator

All Implemented Interfaces:

QuerySegmentWalker, Appenderator

public class BatchAppenderator
extends Object
implements Appenderator

This is a new class produced when the old AppenderatorImpl was split. For historical reasons, the code for creating segments was all handled by the same code path in that class. The code was correct but inefficient for batch ingestion from a memory perspective. If the input file being processed by batch ingestion had enough sinks & hydrants produced then it may run out of memory either in the hydrant creation phase (append) of this class or in the merge hydrants phase. Therefore, a new class, BatchAppenderator, this class, was created to specialize in batch ingestion and the old class for stream ingestion was renamed to StreamAppenderator.

Nested Class Summary

Nested classes/interfaces inherited from interface org.apache.druid.segment.realtime.appenderator.Appenderator

Appenderator.AppenderatorAddResult

Field Summary

Fields

Modifier and Type	Field and Description
static int	ROUGH_OVERHEAD_PER_HYDRANT
static int	ROUGH_OVERHEAD_PER_SINK

Method Summary

Modifier and Type	Method and Description
Appenderator.AppenderatorAddResult	add(SegmentIdWithShardSpec identifier, InputRow row, com.google.common.base.Supplier<Committer> committerSupplier, boolean allowIncrementalPersists) Add a row.
void	clear() Drop all in-memory and on-disk data, and forget any previously-remembered commit metadata.
void	close() Stop any currently-running processing and clean up after ourselves.
void	closeNow() Stop all processing, abandoning current pushes, currently running persist may be allowed to finish if they persist critical metadata otherwise shutdown immediately.
com.google.common.util.concurrent.ListenableFuture<?>	drop(SegmentIdWithShardSpec identifier) Schedule dropping all data associated with a particular pending segment.
long	getBytesCurrentlyInMemory()
long	getBytesInMemory(SegmentIdWithShardSpec identifier)
String	getDataSource() Return the name of the dataSource associated with this Appenderator.
String	getId() Return the identifier of this Appenderator; useful for log messages and such.
List<SegmentIdWithShardSpec>	getInMemorySegments()
List<File>	getPersistedidentifierPaths()
<T> QueryRunner<T>	getQueryRunnerForIntervals(Query<T> query, Iterable<org.joda.time.Interval> intervals)
<T> QueryRunner<T>	getQueryRunnerForSegments(Query<T> query, Iterable<SegmentDescriptor> specs)
int	getRowCount(SegmentIdWithShardSpec identifier) Returns the number of rows in a particular pending segment.
int	getRowsInMemory()
List<SegmentIdWithShardSpec>	getSegments() Returns all active segments regardless whether they are in memory or persisted
int	getTotalRowCount() Returns the number of total rows in this appenderator of all segments pending push.
com.google.common.util.concurrent.ListenableFuture<Object>	persistAll(Committer committer) Persist any in-memory indexed data to durable storage.
com.google.common.util.concurrent.ListenableFuture<SegmentsAndCommitMetadata>	push(Collection<SegmentIdWithShardSpec> identifiers, Committer committer, boolean useUniquePath) Merge and push particular segments to deep storage.
Object	startJob() Perform any initial setup.

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.apache.druid.segment.realtime.appenderator.Appenderator

add

Field Detail

ROUGH_OVERHEAD_PER_SINK

public static final int ROUGH_OVERHEAD_PER_SINK

See Also:

Constant Field Values

ROUGH_OVERHEAD_PER_HYDRANT

public static final int ROUGH_OVERHEAD_PER_HYDRANT

See Also:

Constant Field Values

Method Detail

getId

public String getId()

Description copied from interface: Appenderator

Return the identifier of this Appenderator; useful for log messages and such.

Specified by:

getId in interface Appenderator

getDataSource

public String getDataSource()

Description copied from interface: Appenderator

Return the name of the dataSource associated with this Appenderator.

Specified by:

getDataSource in interface Appenderator

startJob

public Object startJob()

Description copied from interface: Appenderator

Perform any initial setup. Should be called before using any other methods.

Specified by:

startJob in interface Appenderator

Returns:

currently persisted commit metadata

add

public Appenderator.AppenderatorAddResult add(SegmentIdWithShardSpec identifier,
                                              InputRow row,
                                              @Nullable
                                              com.google.common.base.Supplier<Committer> committerSupplier,
                                              boolean allowIncrementalPersists)
                                       throws IndexSizeExceededException,
                                              SegmentNotWritableException

Description copied from interface: Appenderator

Add a row. Must not be called concurrently from multiple threads.

If no pending segment exists for the provided identifier, a new one will be created.

This method may trigger a Appenderator.persistAll(Committer) using the supplied Committer. If it does this, the Committer is guaranteed to be *created* synchronously with the call to add, but will actually be used asynchronously.

If committer is not provided, no metadata is persisted.

Specified by:

add in interface Appenderator

Parameters:

identifier - the segment into which this row should be added

row - the row to add

committerSupplier - supplier of a committer associated with all data that has been added, including this row if allowIncrementalPersists is set to false then this will not be used as no persist will be done automatically

allowIncrementalPersists - indicate whether automatic persist should be performed or not if required. If this flag is set to false then the return value should have Appenderator.AppenderatorAddResult.isPersistRequired set to true if persist was skipped because of this flag and it is assumed that the responsibility of calling Appenderator.persistAll(Committer) is on the caller.

Returns:

Appenderator.AppenderatorAddResult

Throws:

IndexSizeExceededException - if this row cannot be added because it is too large

SegmentNotWritableException - if the requested segment is known, but has been closed

getSegments

public List<SegmentIdWithShardSpec> getSegments()

Returns all active segments regardless whether they are in memory or persisted

Specified by:

getSegments in interface Appenderator

getInMemorySegments

public List<SegmentIdWithShardSpec> getInMemorySegments()

getRowCount

public int getRowCount(SegmentIdWithShardSpec identifier)

Description copied from interface: Appenderator

Returns the number of rows in a particular pending segment.

Specified by:

getRowCount in interface Appenderator

Parameters:

identifier - segment to examine

Returns:

row count

getTotalRowCount

public int getTotalRowCount()

Description copied from interface: Appenderator

Returns the number of total rows in this appenderator of all segments pending push.

Specified by:

getTotalRowCount in interface Appenderator

Returns:

total number of rows

getRowsInMemory

public int getRowsInMemory()

getBytesCurrentlyInMemory

public long getBytesCurrentlyInMemory()

getBytesInMemory

public long getBytesInMemory(SegmentIdWithShardSpec identifier)

getQueryRunnerForIntervals

public <T> QueryRunner<T> getQueryRunnerForIntervals(Query<T> query,
                                                     Iterable<org.joda.time.Interval> intervals)

Specified by:

getQueryRunnerForIntervals in interface QuerySegmentWalker

getQueryRunnerForSegments

public <T> QueryRunner<T> getQueryRunnerForSegments(Query<T> query,
                                                    Iterable<SegmentDescriptor> specs)

Specified by:

getQueryRunnerForSegments in interface QuerySegmentWalker

clear

public void clear()

Description copied from interface: Appenderator

Drop all in-memory and on-disk data, and forget any previously-remembered commit metadata. This could be useful if, for some reason, rows have been added that we do not actually want to hand off. Blocks until all data has been cleared. This may take some time, since all pending persists must finish first.

Specified by:

clear in interface Appenderator

drop

public com.google.common.util.concurrent.ListenableFuture<?> drop(SegmentIdWithShardSpec identifier)

Description copied from interface: Appenderator

Schedule dropping all data associated with a particular pending segment. Unlike Appenderator.clear()), any on-disk commit metadata will remain unchanged. If there is no pending segment with this identifier, then this method will do nothing.

You should not write to the dropped segment after calling "drop". If you need to drop all your data and re-write it, consider Appenderator.clear() instead. This method might be called concurrently from a thread different from the "main data appending / indexing thread", from where all other methods in this class (except those inherited from QuerySegmentWalker) are called. This typically happens when drop() is called in an async future callback. drop() itself is cheap and relays heavy dropping work to an internal executor of this Appenderator.

Specified by:

drop in interface Appenderator

Parameters:

identifier - the pending segment to drop

Returns:

future that resolves when data is dropped

persistAll

public com.google.common.util.concurrent.ListenableFuture<Object> persistAll(@Nullable
                                                                             Committer committer)

Description copied from interface: Appenderator

Persist any in-memory indexed data to durable storage. This may be only somewhat durable, e.g. the machine's local disk. The Committer will be made synchronously with the call to persistAll, but will actually be used asynchronously. Any metadata returned by the committer will be associated with the data persisted to disk.

If committer is not provided, no metadata is persisted.

Specified by:

persistAll in interface Appenderator

Parameters:

committer - a committer associated with all data that has been added so far

Returns:

future that resolves when all pending data has been persisted, contains commit metadata for this persist

push

public com.google.common.util.concurrent.ListenableFuture<SegmentsAndCommitMetadata> push(Collection<SegmentIdWithShardSpec> identifiers,
                                                                                          @Nullable
                                                                                          Committer committer,
                                                                                          boolean useUniquePath)

Description copied from interface: Appenderator

Merge and push particular segments to deep storage. This will trigger an implicit Appenderator.persistAll(Committer) using the provided Committer.

After this method is called, you cannot add new data to any segments that were previously under construction.

If committer is not provided, no metadata is persisted.

Specified by:

push in interface Appenderator

Parameters:

identifiers - list of segments to push

committer - a committer associated with all data that has been added so far

useUniquePath - true if the segment should be written to a path with a unique identifier

Returns:

future that resolves when all segments have been pushed. The segment list will be the list of segments that have been pushed and the commit metadata from the Committer.

close

public void close()

Description copied from interface: Appenderator

Stop any currently-running processing and clean up after ourselves. This allows currently running persists and pushes to finish. This will not remove any on-disk persisted data, but it will drop any data that has not yet been persisted.

Specified by:

close in interface Appenderator

closeNow

public void closeNow()

Description copied from interface: Appenderator

Stop all processing, abandoning current pushes, currently running persist may be allowed to finish if they persist critical metadata otherwise shutdown immediately. This will not remove any on-disk persisted data, but it will drop any data that has not yet been persisted. Since this does not wait for pushes to finish, implementations have to make sure if any push is still happening in background thread then it does not cause any problems.

Specified by:

closeNow in interface Appenderator

getPersistedidentifierPaths

@Nullable
public List<File> getPersistedidentifierPaths()