org.apache.bookkeeper.client

Class RegionAwareEnsemblePlacementPolicy

java.lang.Object

org.apache.bookkeeper.client.TopologyAwareEnsemblePlacementPolicy

org.apache.bookkeeper.client.RackawareEnsemblePlacementPolicyImpl

org.apache.bookkeeper.client.RackawareEnsemblePlacementPolicy

org.apache.bookkeeper.client.RegionAwareEnsemblePlacementPolicy

All Implemented Interfaces:

EnsemblePlacementPolicy, ITopologyAwareEnsemblePlacementPolicy<org.apache.bookkeeper.net.BookieNode>

public class RegionAwareEnsemblePlacementPolicy
extends RackawareEnsemblePlacementPolicy

A placement policy use region information in the network topology for placing ensembles.

See Also:

EnsemblePlacementPolicy

Nested Class Summary

Nested classes/interfaces inherited from class org.apache.bookkeeper.client.TopologyAwareEnsemblePlacementPolicy

TopologyAwareEnsemblePlacementPolicy.DefaultResolver, TopologyAwareEnsemblePlacementPolicy.DNSResolverDecorator, TopologyAwareEnsemblePlacementPolicy.EnsembleForReplacementWithNoConstraints, TopologyAwareEnsemblePlacementPolicy.RRTopologyAwareCoverageEnsemble, TopologyAwareEnsemblePlacementPolicy.TruePredicate

Nested classes/interfaces inherited from interface org.apache.bookkeeper.client.ITopologyAwareEnsemblePlacementPolicy

ITopologyAwareEnsemblePlacementPolicy.Ensemble<T extends org.apache.bookkeeper.net.Node>, ITopologyAwareEnsemblePlacementPolicy.Predicate<T extends org.apache.bookkeeper.net.Node>

Nested classes/interfaces inherited from interface org.apache.bookkeeper.client.EnsemblePlacementPolicy

EnsemblePlacementPolicy.PlacementPolicyAdherence, EnsemblePlacementPolicy.PlacementResult<T>

Field Summary

Fields

Modifier and Type	Field and Description
protected java.util.concurrent.ConcurrentMap<org.apache.bookkeeper.net.BookieSocketAddress,java.lang.String>	address2Region
protected Feature	disableDurabilityFeature
protected java.lang.String	disallowBookiePlacementInRegionFeatureName
protected boolean	enableValidation
protected boolean	enforceDurabilityInReplace
protected FeatureProvider	featureProvider
(package private) static org.slf4j.Logger	LOG
(package private) static int	MINIMUM_REGIONS_FOR_DURABILITY_DEFAULT
protected int	minRegionsForDurability
protected java.lang.String	myRegion
protected java.util.Map<java.lang.String,TopologyAwareEnsemblePlacementPolicy>	perRegionPlacement
static java.lang.String	REGION_AWARE_ANOMALOUS_ENSEMBLE
(package private) static int	REGIONID_DISTANCE_FROM_LEAVES
(package private) static int	REMOTE_NODE_IN_REORDER_SEQUENCE
static java.lang.String	REPP_DISABLE_DURABILITY_ENFORCEMENT_FEATURE
static java.lang.String	REPP_DISABLE_DURABILITY_FEATURE_NAME
static java.lang.String	REPP_DISALLOW_BOOKIE_PLACEMENT_IN_REGION_FEATURE_NAME
static java.lang.String	REPP_ENABLE_DURABILITY_ENFORCEMENT_IN_REPLACE
static java.lang.String	REPP_ENABLE_VALIDATION
static java.lang.String	REPP_MINIMUM_REGIONS_FOR_DURABILITY
static java.lang.String	REPP_REGIONS_TO_WRITE
(package private) static java.lang.String	UNKNOWN_REGION

Fields inherited from class org.apache.bookkeeper.client.RackawareEnsemblePlacementPolicy

slave

Fields inherited from class org.apache.bookkeeper.client.RackawareEnsemblePlacementPolicyImpl

enforceDurability, enforceMinNumRacksPerWriteQuorum, failedToResolveNetworkLocationCounter, ignoreLocalNodeInPlacementPolicy, LOCAL_FAIL_MASK, LOCAL_MASK, localNode, MASK_BITS, maxWeightMultiple, minNumRacksPerWriteQuorum, numWritableBookiesInDefaultRack, RACKNAME_DISTANCE_FROM_LEAVES, READ_ONLY_MASK, readReorderedCounter, REMOTE_FAIL_MASK, REMOTE_MASK, reorderReadsRandom, reorderThresholdPendingRequests, REPP_DNS_RESOLVER_CLASS, REPP_RANDOM_READ_REORDERING, SLOW_MASK, slowBookies, stabilizePeriodSeconds, statsLogger, timer, UNAVAIL_MASK

Fields inherited from class org.apache.bookkeeper.client.TopologyAwareEnsemblePlacementPolicy

bookieInfoMap, bookiesJoinedCounter, bookiesLeftCounter, dnsResolver, isWeighted, knownBookies, readOnlyBookies, rwLock, topology, weightedSelection

Constructor Summary

Constructors

Constructor and Description
RegionAwareEnsemblePlacementPolicy()

Method Summary

Modifier and Type	Method and Description
protected java.lang.String	getLocalRegion(org.apache.bookkeeper.net.BookieNode node)
protected java.lang.String	getRegion(org.apache.bookkeeper.net.BookieSocketAddress addr)
void	handleBookiesThatJoined(java.util.Set<org.apache.bookkeeper.net.BookieSocketAddress> joinedBookies) Handle bookies that joined.
void	handleBookiesThatLeft(java.util.Set<org.apache.bookkeeper.net.BookieSocketAddress> leftBookies) Handle bookies that left.
RegionAwareEnsemblePlacementPolicy	initialize(ClientConfiguration conf, java.util.Optional<org.apache.bookkeeper.net.DNSToSwitchMapping> optionalDnsResolver, io.netty.util.HashedWheelTimer timer, FeatureProvider featureProvider, StatsLogger statsLogger) Initialize the policy.
EnsemblePlacementPolicy.PlacementPolicyAdherence	isEnsembleAdheringToPlacementPolicy(java.util.List<org.apache.bookkeeper.net.BookieSocketAddress> ensembleList, int writeQuorumSize, int ackQuorumSize) returns AdherenceLevel if the Ensemble is strictly/softly/fails adhering to placement policy, like in the case of RackawareEnsemblePlacementPolicy, bookies in the writeset are from 'minNumRacksPerWriteQuorum' number of racks.
EnsemblePlacementPolicy.PlacementResult<java.util.List<org.apache.bookkeeper.net.BookieSocketAddress>>	newEnsemble(int ensembleSize, int writeQuorumSize, int ackQuorumSize, java.util.Map<java.lang.String,byte[]> customMetadata, java.util.Set<org.apache.bookkeeper.net.BookieSocketAddress> excludedBookies) Choose numBookies bookies for ensemble.
DistributionSchedule.WriteSet	reorderReadLACSequence(java.util.List<org.apache.bookkeeper.net.BookieSocketAddress> ensemble, BookiesHealthInfo bookiesHealthInfo, DistributionSchedule.WriteSet writeSet) Reorder the read last add confirmed sequence of a given write quorum writeSet.
DistributionSchedule.WriteSet	reorderReadSequence(java.util.List<org.apache.bookkeeper.net.BookieSocketAddress> ensemble, BookiesHealthInfo bookiesHealthInfo, DistributionSchedule.WriteSet writeSet) Reorder the read sequence of a given write quorum writeSet.
EnsemblePlacementPolicy.PlacementResult<org.apache.bookkeeper.net.BookieSocketAddress>	replaceBookie(int ensembleSize, int writeQuorumSize, int ackQuorumSize, java.util.Map<java.lang.String,byte[]> customMetadata, java.util.List<org.apache.bookkeeper.net.BookieSocketAddress> currentEnsemble, org.apache.bookkeeper.net.BookieSocketAddress bookieToReplace, java.util.Set<org.apache.bookkeeper.net.BookieSocketAddress> excludeBookies) Choose a new bookie to replace bookieToReplace.
protected org.apache.bookkeeper.net.BookieNode	replaceFromRack(org.apache.bookkeeper.net.BookieNode bookieNodeToReplace, java.util.Set<org.apache.bookkeeper.net.Node> excludeBookies, ITopologyAwareEnsemblePlacementPolicy.Predicate<org.apache.bookkeeper.net.BookieNode> predicate, ITopologyAwareEnsemblePlacementPolicy.Ensemble<org.apache.bookkeeper.net.BookieNode> ensemble, boolean enforceDurability)
protected java.util.List<org.apache.bookkeeper.net.BookieNode>	selectRandomFromRegions(java.util.Set<java.lang.String> availableRegions, int numBookies, java.util.Set<org.apache.bookkeeper.net.Node> excludeBookies, ITopologyAwareEnsemblePlacementPolicy.Predicate<org.apache.bookkeeper.net.BookieNode> predicate, ITopologyAwareEnsemblePlacementPolicy.Ensemble<org.apache.bookkeeper.net.BookieNode> ensemble)

Methods inherited from class org.apache.bookkeeper.client.RackawareEnsemblePlacementPolicy

initialize, newEnsemble, onClusterChanged, selectFromNetworkLocation, selectFromNetworkLocation, selectFromNetworkLocation, uninitalize

Methods inherited from class org.apache.bookkeeper.client.RackawareEnsemblePlacementPolicyImpl

addDefaultRackBookiesIfMinNumRacksIsEnforced, areAckedBookiesAdheringToPlacementPolicy, getDefaultRack, newEnsembleInternal, registerSlowBookie, reorderReadSequenceWithRegion, selectRandom, selectRandomFromRack, selectRandomInternal, withDefaultRack

Methods inherited from class org.apache.bookkeeper.client.TopologyAwareEnsemblePlacementPolicy

convertBookiesToNodes, createBookieNode, getNetworkLocations, onBookieRackChange, resolveNetworkLocation, shuffleWithMask, updateBookieInfo

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.apache.bookkeeper.client.ITopologyAwareEnsemblePlacementPolicy

onBookieRackChange

Methods inherited from interface org.apache.bookkeeper.client.EnsemblePlacementPolicy

areAckedBookiesAdheringToPlacementPolicy, getStickyReadBookieIndex, registerSlowBookie, updateBookieInfo

Field Detail

LOG

static final org.slf4j.Logger LOG

REPP_REGIONS_TO_WRITE

public static final java.lang.String REPP_REGIONS_TO_WRITE

See Also:

Constant Field Values

REPP_MINIMUM_REGIONS_FOR_DURABILITY

public static final java.lang.String REPP_MINIMUM_REGIONS_FOR_DURABILITY

See Also:

Constant Field Values

REPP_ENABLE_DURABILITY_ENFORCEMENT_IN_REPLACE

public static final java.lang.String REPP_ENABLE_DURABILITY_ENFORCEMENT_IN_REPLACE

See Also:

Constant Field Values

REPP_DISABLE_DURABILITY_FEATURE_NAME

public static final java.lang.String REPP_DISABLE_DURABILITY_FEATURE_NAME

See Also:

Constant Field Values

REPP_DISALLOW_BOOKIE_PLACEMENT_IN_REGION_FEATURE_NAME

public static final java.lang.String REPP_DISALLOW_BOOKIE_PLACEMENT_IN_REGION_FEATURE_NAME

See Also:

Constant Field Values

REPP_DISABLE_DURABILITY_ENFORCEMENT_FEATURE

public static final java.lang.String REPP_DISABLE_DURABILITY_ENFORCEMENT_FEATURE

See Also:

Constant Field Values

REPP_ENABLE_VALIDATION

public static final java.lang.String REPP_ENABLE_VALIDATION

See Also:

Constant Field Values

REGION_AWARE_ANOMALOUS_ENSEMBLE

public static final java.lang.String REGION_AWARE_ANOMALOUS_ENSEMBLE

See Also:

Constant Field Values

MINIMUM_REGIONS_FOR_DURABILITY_DEFAULT

static final int MINIMUM_REGIONS_FOR_DURABILITY_DEFAULT

See Also:

Constant Field Values

REGIONID_DISTANCE_FROM_LEAVES

static final int REGIONID_DISTANCE_FROM_LEAVES

See Also:

Constant Field Values

UNKNOWN_REGION

static final java.lang.String UNKNOWN_REGION

See Also:

Constant Field Values

REMOTE_NODE_IN_REORDER_SEQUENCE

static final int REMOTE_NODE_IN_REORDER_SEQUENCE

See Also:

Constant Field Values

perRegionPlacement

protected final java.util.Map<java.lang.String,TopologyAwareEnsemblePlacementPolicy> perRegionPlacement

address2Region

protected final java.util.concurrent.ConcurrentMap<org.apache.bookkeeper.net.BookieSocketAddress,java.lang.String> address2Region

featureProvider

protected FeatureProvider featureProvider

disallowBookiePlacementInRegionFeatureName

protected java.lang.String disallowBookiePlacementInRegionFeatureName

myRegion

protected java.lang.String myRegion

minRegionsForDurability

protected int minRegionsForDurability

enableValidation

protected boolean enableValidation

enforceDurabilityInReplace

protected boolean enforceDurabilityInReplace

disableDurabilityFeature

protected Feature disableDurabilityFeature

Constructor Detail

RegionAwareEnsemblePlacementPolicy

RegionAwareEnsemblePlacementPolicy()

Method Detail

getRegion

protected java.lang.String getRegion(org.apache.bookkeeper.net.BookieSocketAddress addr)

getLocalRegion

protected java.lang.String getLocalRegion(org.apache.bookkeeper.net.BookieNode node)

handleBookiesThatLeft

public void handleBookiesThatLeft(java.util.Set<org.apache.bookkeeper.net.BookieSocketAddress> leftBookies)

Description copied from interface: ITopologyAwareEnsemblePlacementPolicy

Handle bookies that left.

Specified by:

handleBookiesThatLeft in interface ITopologyAwareEnsemblePlacementPolicy<org.apache.bookkeeper.net.BookieNode>

Overrides:

handleBookiesThatLeft in class RackawareEnsemblePlacementPolicy

Parameters:

leftBookies - bookies that left

handleBookiesThatJoined

public void handleBookiesThatJoined(java.util.Set<org.apache.bookkeeper.net.BookieSocketAddress> joinedBookies)

Description copied from interface: ITopologyAwareEnsemblePlacementPolicy

Handle bookies that joined.

Specified by:

handleBookiesThatJoined in interface ITopologyAwareEnsemblePlacementPolicy<org.apache.bookkeeper.net.BookieNode>

Overrides:

handleBookiesThatJoined in class RackawareEnsemblePlacementPolicy

Parameters:

joinedBookies - bookies that joined.

initialize

public RegionAwareEnsemblePlacementPolicy initialize(ClientConfiguration conf,
                                                     java.util.Optional<org.apache.bookkeeper.net.DNSToSwitchMapping> optionalDnsResolver,
                                                     io.netty.util.HashedWheelTimer timer,
                                                     FeatureProvider featureProvider,
                                                     StatsLogger statsLogger)

Description copied from interface: EnsemblePlacementPolicy

Initialize the policy.

Specified by:

initialize in interface EnsemblePlacementPolicy

Overrides:

initialize in class RackawareEnsemblePlacementPolicyImpl

Parameters:

conf - client configuration

optionalDnsResolver - dns resolver

timer - timer

featureProvider - feature provider

statsLogger - stats logger

selectRandomFromRegions

protected java.util.List<org.apache.bookkeeper.net.BookieNode> selectRandomFromRegions(java.util.Set<java.lang.String> availableRegions,
                                                                                       int numBookies,
                                                                                       java.util.Set<org.apache.bookkeeper.net.Node> excludeBookies,
                                                                                       ITopologyAwareEnsemblePlacementPolicy.Predicate<org.apache.bookkeeper.net.BookieNode> predicate,
                                                                                       ITopologyAwareEnsemblePlacementPolicy.Ensemble<org.apache.bookkeeper.net.BookieNode> ensemble)
                                                                                throws BKException.BKNotEnoughBookiesException

Throws:

BKException.BKNotEnoughBookiesException

newEnsemble

public EnsemblePlacementPolicy.PlacementResult<java.util.List<org.apache.bookkeeper.net.BookieSocketAddress>> newEnsemble(int ensembleSize,
                                                                                                                          int writeQuorumSize,
                                                                                                                          int ackQuorumSize,
                                                                                                                          java.util.Map<java.lang.String,byte[]> customMetadata,
                                                                                                                          java.util.Set<org.apache.bookkeeper.net.BookieSocketAddress> excludedBookies)
                                                                                                                   throws BKException.BKNotEnoughBookiesException

Description copied from interface: EnsemblePlacementPolicy

Choose numBookies bookies for ensemble. If the count is more than the number of available nodes, BKException.BKNotEnoughBookiesException is thrown.

The implementation should respect to the replace settings. The size of the returned bookie list should be equal to the provide ensembleSize.

customMetadata is the same user defined data that user provides when BookKeeper.createLedger(int, int, int, BookKeeper.DigestType, byte[], Map).

If 'enforceMinNumRacksPerWriteQuorum' config is enabled then the bookies belonging to default faultzone (rack) will be excluded while selecting bookies.

Specified by:

newEnsemble in interface EnsemblePlacementPolicy

Overrides:

newEnsemble in class RackawareEnsemblePlacementPolicy

Parameters:

ensembleSize - Ensemble Size

writeQuorumSize - Write Quorum Size

ackQuorumSize - the value of ackQuorumSize (added since 4.5)

customMetadata - the value of customMetadata. it is the same user defined metadata that user provides in BookKeeper.createLedger(int, int, int, BookKeeper.DigestType, byte[])

excludedBookies - Bookies that should not be considered as targets.

Returns:

a placement result containing list of bookie addresses for the ensemble.

Throws:

BKException.BKNotEnoughBookiesException - if not enough bookies available.

replaceBookie

public EnsemblePlacementPolicy.PlacementResult<org.apache.bookkeeper.net.BookieSocketAddress> replaceBookie(int ensembleSize,
                                                                                                            int writeQuorumSize,
                                                                                                            int ackQuorumSize,
                                                                                                            java.util.Map<java.lang.String,byte[]> customMetadata,
                                                                                                            java.util.List<org.apache.bookkeeper.net.BookieSocketAddress> currentEnsemble,
                                                                                                            org.apache.bookkeeper.net.BookieSocketAddress bookieToReplace,
                                                                                                            java.util.Set<org.apache.bookkeeper.net.BookieSocketAddress> excludeBookies)
                                                                                                     throws BKException.BKNotEnoughBookiesException

Description copied from interface: EnsemblePlacementPolicy

Choose a new bookie to replace bookieToReplace. If no bookie available in the cluster, BKException.BKNotEnoughBookiesException is thrown.

If 'enforceMinNumRacksPerWriteQuorum' config is enabled then the bookies belonging to default faultzone (rack) will be excluded while selecting bookies.

Specified by:

replaceBookie in interface EnsemblePlacementPolicy

Overrides:

replaceBookie in class RackawareEnsemblePlacementPolicy

Parameters:

ensembleSize - the value of ensembleSize

writeQuorumSize - the value of writeQuorumSize

ackQuorumSize - the value of ackQuorumSize (added since 4.5)

customMetadata - the value of customMetadata. it is the same user defined metadata that user provides in BookKeeper.createLedger(int, int, int, BookKeeper.DigestType, byte[])

currentEnsemble - the value of currentEnsemble

bookieToReplace - bookie to replace

excludeBookies - bookies that should not be considered as candidate.

Returns:

a placement result containing the new bookie address.

Throws:

BKException.BKNotEnoughBookiesException

replaceFromRack

protected org.apache.bookkeeper.net.BookieNode replaceFromRack(org.apache.bookkeeper.net.BookieNode bookieNodeToReplace,
                                                               java.util.Set<org.apache.bookkeeper.net.Node> excludeBookies,
                                                               ITopologyAwareEnsemblePlacementPolicy.Predicate<org.apache.bookkeeper.net.BookieNode> predicate,
                                                               ITopologyAwareEnsemblePlacementPolicy.Ensemble<org.apache.bookkeeper.net.BookieNode> ensemble,
                                                               boolean enforceDurability)
                                                        throws BKException.BKNotEnoughBookiesException

Throws:

BKException.BKNotEnoughBookiesException

reorderReadSequence

public final DistributionSchedule.WriteSet reorderReadSequence(java.util.List<org.apache.bookkeeper.net.BookieSocketAddress> ensemble,
                                                               BookiesHealthInfo bookiesHealthInfo,
                                                               DistributionSchedule.WriteSet writeSet)

Description copied from interface: EnsemblePlacementPolicy

Reorder the read sequence of a given write quorum writeSet.

Specified by:

reorderReadSequence in interface EnsemblePlacementPolicy

Overrides:

reorderReadSequence in class RackawareEnsemblePlacementPolicy

Parameters:

ensemble - Ensemble to read entries.

bookiesHealthInfo - Health info for bookies

writeSet - Write quorum to read entries. This will be modified, rather than allocating a new WriteSet.

Returns:

The read sequence. This will be the same object as the passed in writeSet.

reorderReadLACSequence

public final DistributionSchedule.WriteSet reorderReadLACSequence(java.util.List<org.apache.bookkeeper.net.BookieSocketAddress> ensemble,
                                                                  BookiesHealthInfo bookiesHealthInfo,
                                                                  DistributionSchedule.WriteSet writeSet)

Description copied from interface: EnsemblePlacementPolicy

Reorder the read last add confirmed sequence of a given write quorum writeSet.

Specified by:

reorderReadLACSequence in interface EnsemblePlacementPolicy

Overrides:

reorderReadLACSequence in class RackawareEnsemblePlacementPolicy

Parameters:

ensemble - Ensemble to read entries.

bookiesHealthInfo - Health info for bookies

writeSet - Write quorum to read entries. This will be modified, rather than allocating a new WriteSet.

Returns:

The read sequence. This will be the same object as the passed in writeSet.

isEnsembleAdheringToPlacementPolicy

public EnsemblePlacementPolicy.PlacementPolicyAdherence isEnsembleAdheringToPlacementPolicy(java.util.List<org.apache.bookkeeper.net.BookieSocketAddress> ensembleList,
                                                                                            int writeQuorumSize,
                                                                                            int ackQuorumSize)

Description copied from interface: EnsemblePlacementPolicy

returns AdherenceLevel if the Ensemble is strictly/softly/fails adhering to placement policy, like in the case of RackawareEnsemblePlacementPolicy, bookies in the writeset are from 'minNumRacksPerWriteQuorum' number of racks. And in the case of RegionawareEnsemblePlacementPolicy, check for minimumRegionsForDurability, reppRegionsToWrite, rack distribution within a region and other parameters of RegionAwareEnsemblePlacementPolicy. In ZoneAwareEnsemblePlacementPolicy if bookies in the writeset are from 'desiredNumOfZones' then it is considered as MEETS_STRICT if they are from 'minNumOfZones' then it is considered as MEETS_SOFT otherwise considered as FAIL.

Specified by:

isEnsembleAdheringToPlacementPolicy in interface EnsemblePlacementPolicy

Overrides:

isEnsembleAdheringToPlacementPolicy in class RackawareEnsemblePlacementPolicyImpl

Parameters:

ensembleList - list of BookieSocketAddress of bookies in the ensemble

writeQuorumSize - writeQuorumSize of the ensemble

ackQuorumSize - ackQuorumSize of the ensemble

Returns: