org.apache.hadoop.hbase.regionserver.compactions

类 SortedCompactionPolicy

java.lang.Object

org.apache.hadoop.hbase.regionserver.compactions.CompactionPolicy

org.apache.hadoop.hbase.regionserver.compactions.SortedCompactionPolicy

直接已知子类:

DateTieredCompactionPolicy, RatioBasedCompactionPolicy

@InterfaceAudience.Private
public abstract class SortedCompactionPolicy
extends CompactionPolicy

An abstract compaction policy that select files on seq id order.

字段概要

从类继承的字段 org.apache.hadoop.hbase.regionserver.compactions.CompactionPolicy

comConf, storeConfigInfo

构造器概要

构造器

构造器和说明
SortedCompactionPolicy(org.apache.hadoop.conf.Configuration conf, StoreConfigInformation storeConfigInfo)

方法概要

限定符和类型	方法和说明
protected ArrayList<HStoreFile>	checkMinFilesCriteria(ArrayList<HStoreFile> candidates, int minFiles)
protected abstract CompactionRequestImpl	createCompactionRequest(ArrayList<HStoreFile> candidateSelection, boolean tryingMajor, boolean mayUseOffPeak, boolean mayBeStuck)
protected ArrayList<HStoreFile>	filterBulk(ArrayList<HStoreFile> candidates)
protected ArrayList<HStoreFile>	getCurrentEligibleFiles(ArrayList<HStoreFile> candidateFiles, List<HStoreFile> filesCompacting)
long	getNextMajorCompactTime(Collection<HStoreFile> filesToCompact)
abstract boolean	needsCompaction(Collection<HStoreFile> storeFiles, List<HStoreFile> filesCompacting)
List<HStoreFile>	preSelectCompactionForCoprocessor(Collection<HStoreFile> candidates, List<HStoreFile> filesCompacting)
protected void	removeExcessFiles(ArrayList<HStoreFile> candidates, boolean isUserCompaction, boolean isMajorCompaction)
CompactionRequestImpl	selectCompaction(Collection<HStoreFile> candidateFiles, List<HStoreFile> filesCompacting, boolean isUserCompaction, boolean mayUseOffPeak, boolean forceMajor)
abstract boolean	shouldPerformMajorCompaction(Collection<HStoreFile> filesToCompact)
protected ArrayList<HStoreFile>	skipLargeFiles(ArrayList<HStoreFile> candidates, boolean mayUseOffpeak)
boolean	throttleCompaction(long compactionSize)

从类继承的方法 org.apache.hadoop.hbase.regionserver.compactions.CompactionPolicy

getConf, setConf

从类继承的方法 java.lang.Object

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

构造器详细资料

SortedCompactionPolicy

public SortedCompactionPolicy(org.apache.hadoop.conf.Configuration conf,
                              StoreConfigInformation storeConfigInfo)

方法详细资料

preSelectCompactionForCoprocessor

public List<HStoreFile> preSelectCompactionForCoprocessor(Collection<HStoreFile> candidates,
                                                          List<HStoreFile> filesCompacting)

selectCompaction

public CompactionRequestImpl selectCompaction(Collection<HStoreFile> candidateFiles,
                                              List<HStoreFile> filesCompacting,
                                              boolean isUserCompaction,
                                              boolean mayUseOffPeak,
                                              boolean forceMajor)
                                       throws IOException

参数:

candidateFiles - candidate files, ordered from oldest to newest by seqId. We rely on DefaultStoreFileManager to sort the files by seqId to guarantee contiguous compaction based on seqId for data consistency.

返回:

subset copy of candidate list that meets compaction criteria

抛出:

IOException

createCompactionRequest

protected abstract CompactionRequestImpl createCompactionRequest(ArrayList<HStoreFile> candidateSelection,
                                                                 boolean tryingMajor,
                                                                 boolean mayUseOffPeak,
                                                                 boolean mayBeStuck)
                                                          throws IOException

抛出:

IOException

shouldPerformMajorCompaction

public abstract boolean shouldPerformMajorCompaction(Collection<HStoreFile> filesToCompact)
                                              throws IOException

指定者:

shouldPerformMajorCompaction 在类中 CompactionPolicy

参数:

filesToCompact - Files to compact. Can be null.

返回:

True if we should run a major compaction.

抛出:

IOException

getNextMajorCompactTime

public long getNextMajorCompactTime(Collection<HStoreFile> filesToCompact)

参数:

filesToCompact -

返回:

When to run next major compaction

throttleCompaction

public boolean throttleCompaction(long compactionSize)

指定者:

throttleCompaction 在类中 CompactionPolicy

参数:

compactionSize - Total size of some compaction

返回:

whether this should be a large or small compaction

needsCompaction

public abstract boolean needsCompaction(Collection<HStoreFile> storeFiles,
                                        List<HStoreFile> filesCompacting)

getCurrentEligibleFiles

protected ArrayList<HStoreFile> getCurrentEligibleFiles(ArrayList<HStoreFile> candidateFiles,
                                                        List<HStoreFile> filesCompacting)

skipLargeFiles

protected ArrayList<HStoreFile> skipLargeFiles(ArrayList<HStoreFile> candidates,
                                               boolean mayUseOffpeak)

参数:

candidates - pre-filtrate

返回:

filtered subset exclude all files above maxCompactSize Also save all references. We MUST compact them

filterBulk

protected ArrayList<HStoreFile> filterBulk(ArrayList<HStoreFile> candidates)

参数:

candidates - pre-filtrate

返回:

filtered subset exclude all bulk load files if configured

removeExcessFiles

protected void removeExcessFiles(ArrayList<HStoreFile> candidates,
                                 boolean isUserCompaction,
                                 boolean isMajorCompaction)

参数:

candidates - pre-filtrate

checkMinFilesCriteria

protected ArrayList<HStoreFile> checkMinFilesCriteria(ArrayList<HStoreFile> candidates,
                                                      int minFiles)

参数:

candidates - pre-filtrate

返回:

filtered subset forget the compactionSelection if we don't have enough files