org.apache.hadoop.hbase.regionserver

Class StoreEngine<SF extends StoreFlusher,CP extends CompactionPolicy,C extends Compactor<?>,SFM extends StoreFileManager>

java.lang.Object

org.apache.hadoop.hbase.regionserver.StoreEngine<SF,CP,C,SFM>

Direct Known Subclasses:

DateTieredStoreEngine, DefaultStoreEngine, StripeStoreEngine

@InterfaceAudience.Private
public abstract class StoreEngine<SF extends StoreFlusher,CP extends CompactionPolicy,C extends Compactor<?>,SFM extends StoreFileManager>
extends Object

StoreEngine is a factory that can create the objects necessary for HStore to operate. Since not all compaction policies, compactors and store file managers are compatible, they are tied together and replaced together via StoreEngine-s.

We expose read write lock methods to upper layer for store operations:

• Locked in shared mode when the list of component stores is looked at:
  • all reads/writes to table data
  • checking for split
• Locked in exclusive mode when the list of component stores is modified:
  • closing
  • completing a compaction

It is a bit confusing that we have a StoreFileManager(SFM) and then a StoreFileTracker(SFT). As its name says, SFT is used to track the store files list. The reason why we have a SFT beside SFM is that, when introducing stripe compaction, we introduced the StoreEngine and also the SFM, but actually, the SFM here is not a general 'Manager', it is only designed to manage the in memory 'stripes', so we can select different store files when scanning or compacting. The 'tracking' of store files is actually done in HRegionFileSystem and HStore before we have SFT. And since SFM is designed to only holds in memory states, we will hold write lock when updating it, the lock is also used to protect the normal read/write requests. This means we'd better not add IO operations to SFM. And also, no matter what the in memory state is, stripe or not, it does not effect how we track the store files. So consider all these facts, here we introduce a separated SFT to track the store files.

Here, since we always need to update SFM and SFT almost at the same time, we introduce methods in StoreEngine directly to update them both, so upper layer just need to update StoreEngine once, to reduce the possible misuse.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static interface	StoreEngine.IOExceptionRunnable

Field Summary

Fields

Modifier and Type	Field and Description
private BloomFilterMetrics	bloomFilterMetrics
protected CP	compactionPolicy
protected C	compactor
private org.apache.hadoop.conf.Configuration	conf
private RegionCoprocessorHost	coprocessorHost
private StoreContext	ctx
private static Class<? extends StoreEngine<?,?,?,?>>	DEFAULT_STORE_ENGINE_CLASS
private static org.slf4j.Logger	LOG
private Function<String,ExecutorService>	openStoreFileThreadPoolCreator
static String	STORE_ENGINE_CLASS_KEY The name of the configuration parameter that specifies the class of a store engine that is used to manage and compact HBase store files.
protected SFM	storeFileManager
private StoreFileTracker	storeFileTracker
protected SF	storeFlusher
private ReadWriteLock	storeLock

Constructor Summary

Constructors

Constructor and Description
StoreEngine()

Method Summary

Modifier and Type	Method and Description
void	addStoreFiles(Collection<HStoreFile> storeFiles, StoreEngine.IOExceptionRunnable actionAfterAdding) Add the store files to store file manager, and also record it in the store file tracker.
List<HStoreFile>	commitStoreFiles(List<org.apache.hadoop.fs.Path> files, boolean validate) Commit the given files.
static StoreEngine<?,?,?,?>	create(HStore store, org.apache.hadoop.conf.Configuration conf, CellComparator cellComparator) Create the StoreEngine configured for the given Store.
abstract CompactionContext	createCompaction() Creates an instance of a compaction context specific to this engine.
protected abstract void	createComponents(org.apache.hadoop.conf.Configuration conf, HStore store, CellComparator cellComparator) Create the StoreEngine's components.
protected void	createComponentsOnce(org.apache.hadoop.conf.Configuration conf, HStore store, CellComparator cellComparator)
HStoreFile	createStoreFileAndReader(org.apache.hadoop.fs.Path p)
HStoreFile	createStoreFileAndReader(StoreFileInfo info)
private StoreFileTracker	createStoreFileTracker(org.apache.hadoop.conf.Configuration conf, HStore store)
StoreFileWriter	createWriter(CreateStoreFileWriterParams params) Create a writer for writing new store files.
BloomFilterMetrics	getBloomFilterMetrics()
CompactionPolicy	getCompactionPolicy() Returns Compaction policy to use.
Compactor<?>	getCompactor() Returns Compactor to use.
(package private) ReadWriteLock	getLock()
StoreFileManager	getStoreFileManager() Returns Store file manager to use.
StoreFlusher	getStoreFlusher() Returns Store flusher to use.
void	initialize(boolean warmup)
abstract boolean	needsCompaction(List<HStoreFile> filesCompacting)
private List<HStoreFile>	openStoreFiles(Collection<StoreFileInfo> files, boolean warmup)
void	readLock() Acquire read lock of this store.
void	readUnlock() Release read lock of this store.
void	refreshStoreFiles()
void	refreshStoreFiles(Collection<String> newFiles)
private void	refreshStoreFilesInternal(Collection<StoreFileInfo> newFiles) Checks the underlying store files, and opens the files that have not been opened, and removes the store file readers for store files no longer available.
void	removeCompactedFiles(Collection<HStoreFile> compactedFiles)
void	replaceStoreFiles(Collection<HStoreFile> compactedFiles, Collection<HStoreFile> newFiles, StoreEngine.IOExceptionRunnable walMarkerWriter, Runnable actionUnderLock)
boolean	requireWritingToTmpDirFirst() Whether the implementation of the used storefile tracker requires you to write to temp directory first, i.e, does not allow broken store files under the actual data directory.
void	validateStoreFile(org.apache.hadoop.fs.Path path) Validates a store file by opening and closing it.
void	writeLock() Acquire write lock of this store.
void	writeUnlock() Release write lock of this store.

Methods inherited from class java.lang.Object

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

Field Detail

LOG

private static final org.slf4j.Logger LOG

storeFlusher

protected SF extends StoreFlusher storeFlusher

compactionPolicy

protected CP extends CompactionPolicy compactionPolicy

compactor

protected C extends Compactor<?> compactor

storeFileManager

protected SFM extends StoreFileManager storeFileManager

bloomFilterMetrics

private final BloomFilterMetrics bloomFilterMetrics

conf

private org.apache.hadoop.conf.Configuration conf

ctx

private StoreContext ctx

coprocessorHost

private RegionCoprocessorHost coprocessorHost

openStoreFileThreadPoolCreator

private Function<String,ExecutorService> openStoreFileThreadPoolCreator

storeFileTracker

private StoreFileTracker storeFileTracker

storeLock

private final ReadWriteLock storeLock

STORE_ENGINE_CLASS_KEY

public static final String STORE_ENGINE_CLASS_KEY

The name of the configuration parameter that specifies the class of a store engine that is used to manage and compact HBase store files.

See Also:

Constant Field Values

DEFAULT_STORE_ENGINE_CLASS

private static final Class<? extends StoreEngine<?,?,?,?>> DEFAULT_STORE_ENGINE_CLASS

Constructor Detail

StoreEngine

public StoreEngine()

Method Detail

readLock

public void readLock()

Acquire read lock of this store.

readUnlock

public void readUnlock()

Release read lock of this store.

writeLock

public void writeLock()

Acquire write lock of this store.

writeUnlock

public void writeUnlock()

Release write lock of this store.

getCompactionPolicy

public CompactionPolicy getCompactionPolicy()

Returns Compaction policy to use.

getCompactor

public Compactor<?> getCompactor()

Returns Compactor to use.

getStoreFileManager

public StoreFileManager getStoreFileManager()

Returns Store file manager to use.

getStoreFlusher

public StoreFlusher getStoreFlusher()

Returns Store flusher to use.

createStoreFileTracker

private StoreFileTracker createStoreFileTracker(org.apache.hadoop.conf.Configuration conf,
                                                HStore store)

needsCompaction

public abstract boolean needsCompaction(List<HStoreFile> filesCompacting)

Parameters:

filesCompacting - Files currently compacting

Returns:

whether a compaction selection is possible

createCompaction

public abstract CompactionContext createCompaction()
                                            throws IOException

Creates an instance of a compaction context specific to this engine. Doesn't actually select or start a compaction. See CompactionContext class comment.

Returns:

New CompactionContext object.

Throws:

IOException

createComponents

protected abstract void createComponents(org.apache.hadoop.conf.Configuration conf,
                                         HStore store,
                                         CellComparator cellComparator)
                                  throws IOException

Create the StoreEngine's components.

Throws:

IOException

createComponentsOnce

protected final void createComponentsOnce(org.apache.hadoop.conf.Configuration conf,
                                          HStore store,
                                          CellComparator cellComparator)
                                   throws IOException

Throws:

IOException

createWriter

public StoreFileWriter createWriter(CreateStoreFileWriterParams params)
                             throws IOException

Create a writer for writing new store files.

Returns:

Writer for a new StoreFile

Throws:

IOException

createStoreFileAndReader

public HStoreFile createStoreFileAndReader(org.apache.hadoop.fs.Path p)
                                    throws IOException

Throws:

IOException

createStoreFileAndReader

public HStoreFile createStoreFileAndReader(StoreFileInfo info)
                                    throws IOException

Throws:

IOException

validateStoreFile

public void validateStoreFile(org.apache.hadoop.fs.Path path)
                       throws IOException

Validates a store file by opening and closing it. In HFileV2 this should not be an expensive operation.

Parameters:

path - the path to the store file

Throws:

IOException

openStoreFiles

private List<HStoreFile> openStoreFiles(Collection<StoreFileInfo> files,
                                        boolean warmup)
                                 throws IOException

Throws:

IOException

initialize

public void initialize(boolean warmup)
                throws IOException

Throws:

IOException

refreshStoreFiles

public void refreshStoreFiles()
                       throws IOException

Throws:

IOException

refreshStoreFiles

public void refreshStoreFiles(Collection<String> newFiles)
                       throws IOException

Throws:

IOException

refreshStoreFilesInternal

private void refreshStoreFilesInternal(Collection<StoreFileInfo> newFiles)
                                throws IOException

Checks the underlying store files, and opens the files that have not been opened, and removes the store file readers for store files no longer available. Mainly used by secondary region replicas to keep up to date with the primary region files.

Throws:

IOException

commitStoreFiles

public List<HStoreFile> commitStoreFiles(List<org.apache.hadoop.fs.Path> files,
                                         boolean validate)
                                  throws IOException

Commit the given files.

We will move the file into data directory, and open it.

Parameters:

files - the files want to commit

validate - whether to validate the store files

Returns:

the committed store files

Throws:

IOException

addStoreFiles

public void addStoreFiles(Collection<HStoreFile> storeFiles,
                          StoreEngine.IOExceptionRunnable actionAfterAdding)
                   throws IOException

Add the store files to store file manager, and also record it in the store file tracker.

The actionAfterAdding will be executed after the insertion to store file manager, under the lock protection. Usually this is for clear the memstore snapshot.

Throws:

IOException

replaceStoreFiles

public void replaceStoreFiles(Collection<HStoreFile> compactedFiles,
                              Collection<HStoreFile> newFiles,
                              StoreEngine.IOExceptionRunnable walMarkerWriter,
                              Runnable actionUnderLock)
                       throws IOException

Throws:

IOException

removeCompactedFiles

public void removeCompactedFiles(Collection<HStoreFile> compactedFiles)

create

public static StoreEngine<?,?,?,?> create(HStore store,
                                          org.apache.hadoop.conf.Configuration conf,
                                          CellComparator cellComparator)
                                   throws IOException

Create the StoreEngine configured for the given Store.

Parameters:

store - The store. An unfortunate dependency needed due to it being passed to coprocessors via the compactor.

conf - Store configuration.

cellComparator - CellComparator for storeFileManager.

Returns:

StoreEngine to use.

Throws:

IOException

requireWritingToTmpDirFirst

public boolean requireWritingToTmpDirFirst()

Whether the implementation of the used storefile tracker requires you to write to temp directory first, i.e, does not allow broken store files under the actual data directory.

getLock

ReadWriteLock getLock()

getBloomFilterMetrics

public BloomFilterMetrics getBloomFilterMetrics()