org.apache.cassandra.db.memtable

Class ShardedSkipListMemtable

java.lang.Object

org.apache.cassandra.db.memtable.AbstractMemtable

org.apache.cassandra.db.memtable.AbstractMemtableWithCommitlog

org.apache.cassandra.db.memtable.AbstractAllocatorMemtable

org.apache.cassandra.db.memtable.ShardedSkipListMemtable

All Implemented Interfaces:

java.lang.Comparable<Memtable>, Memtable, UnfilteredSource

public class ShardedSkipListMemtable
extends AbstractAllocatorMemtable

A proof-of-concept sharded memtable implementation. This implementation splits the partition skip-list into several independent skip-lists each covering a roughly equal part of the token space served by this node. This reduces congestion of the skip-list from concurrent writes and can lead to improved write throughput. The implementation takes two parameters: - shards: the number of shards to split into. - serialize_writes: if false, each shard may serve multiple writes in parallel; if true, writes to each shard are synchronized. Also see Memtable_API.md.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	ShardedSkipListMemtable.MemtableUnfilteredPartitionIterator

Nested classes/interfaces inherited from class org.apache.cassandra.db.memtable.AbstractMemtable

AbstractMemtable.AbstractFlushablePartitionSet<P extends Partition>, AbstractMemtable.ColumnsCollector, AbstractMemtable.StatsCollector

Nested classes/interfaces inherited from interface org.apache.cassandra.db.memtable.Memtable

Memtable.FlushablePartitionSet<P extends Partition>, Memtable.LastCommitLogPosition, Memtable.MemoryUsage, Memtable.Owner

Field Summary

Fields

Modifier and Type	Field and Description
static java.lang.String	LOCKING_OPTION
static java.lang.String	SHARD_COUNT_PROPERTY
static java.lang.String	SHARDS_OPTION

Fields inherited from class org.apache.cassandra.db.memtable.AbstractAllocatorMemtable

allocator, initialComparator, MEMORY_POOL, owner

Fields inherited from class org.apache.cassandra.db.memtable.AbstractMemtable

columnsCollector, currentOperations, metadata, minLocalDeletionTime, minTimestamp, statsCollector

Fields inherited from interface org.apache.cassandra.db.memtable.Memtable

NO_MIN_TIMESTAMP

Method Summary

Modifier and Type	Method and Description
protected Memtable.Factory	factory()
static org.apache.cassandra.db.memtable.ShardedSkipListMemtable.Factory	factory(java.util.Map<java.lang.String,java.lang.String> optionsCopy)
Memtable.FlushablePartitionSet<AtomicBTreePartition>	getFlushSet(PartitionPosition from, PartitionPosition to) Get the collection of data between the given partition boundaries in a form suitable for flushing.
long	getLiveDataSize() Technically we should scatter gather on all the core threads because the size in following calls are not using volatile variables, but for metrics purpose this should be good enough.
int	getMinLocalDeletionTime() Minimum local deletion time in the memtable
long	getMinTimestamp() Returns the minTS if one available, otherwise NO_MIN_TIMESTAMP.
boolean	isClean() True if the memtable contains no data
long	operationCount() Number of "operations" (in the sense defined in PartitionUpdate.operationCount()) the memtable has executed.
long	partitionCount() Number of partitions stored in the memtable
ShardedSkipListMemtable.MemtableUnfilteredPartitionIterator	partitionIterator(ColumnFilter columnFilter, DataRange dataRange, SSTableReadsListener readsListener) Returns a partition iterator for the given data range.
long	put(PartitionUpdate update, UpdateTransaction indexer, OpOrder.Group opGroup) Should only be called by ColumnFamilyStore.apply via Keyspace.apply, which supplies the appropriate OpOrdering.
UnfilteredRowIterator	rowIterator(DecoratedKey key)
UnfilteredRowIterator	rowIterator(DecoratedKey key, Slices slices, ColumnFilter selectedColumns, boolean reversed, SSTableReadsListener listener) Returns a row iterator for the given partition, applying the specified row and column filters.

Methods inherited from class org.apache.cassandra.db.memtable.AbstractAllocatorMemtable

addMemoryUsageTo, createMemtableAllocatorPoolInternal, discard, flushLargestMemtable, getAllocator, localRangesUpdated, markExtraOffHeapUsed, markExtraOnHeapUsed, metadataUpdated, performSnapshot, shouldSwitch, switchOut, toString

Methods inherited from class org.apache.cassandra.db.memtable.AbstractMemtableWithCommitlog

accepts, getApproximateCommitLogLowerBound, getCommitLogLowerBound, getFinalCommitLogUpperBound, mayContainDataBefore

Methods inherited from class org.apache.cassandra.db.memtable.AbstractMemtable

metadata, updateMin, updateMin

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.apache.cassandra.db.memtable.Memtable

compareTo, getMemoryUsage, newMemoryUsage

Field Detail

SHARDS_OPTION

public static final java.lang.String SHARDS_OPTION

See Also:

Constant Field Values

LOCKING_OPTION

public static final java.lang.String LOCKING_OPTION

See Also:

Constant Field Values

SHARD_COUNT_PROPERTY

public static final java.lang.String SHARD_COUNT_PROPERTY

See Also:

Constant Field Values

Method Detail

isClean

public boolean isClean()

Description copied from interface: Memtable

True if the memtable contains no data

factory

protected Memtable.Factory factory()

Specified by:

factory in class AbstractAllocatorMemtable

put

public long put(PartitionUpdate update,
                UpdateTransaction indexer,
                OpOrder.Group opGroup)

Should only be called by ColumnFamilyStore.apply via Keyspace.apply, which supplies the appropriate OpOrdering. commitLogSegmentPosition should only be null if this is a secondary index, in which case it is *expected* to be null

Parameters:

update - the partition update, may be a new partition or an update to an existing one

indexer - receives information about the update's effect

opGroup - write operation group, used to permit the operation to complete if it is needed to complete a flush to free space.

Returns:

the smallest timestamp delta between corresponding rows from existing and update. A timestamp delta being computed as the difference between the cells and DeletionTimes from any existing partition and those in update. See CASSANDRA-7979.

getLiveDataSize

public long getLiveDataSize()

Technically we should scatter gather on all the core threads because the size in following calls are not using volatile variables, but for metrics purpose this should be good enough.

operationCount

public long operationCount()

Description copied from interface: Memtable

Number of "operations" (in the sense defined in PartitionUpdate.operationCount()) the memtable has executed.

Specified by:

operationCount in interface Memtable

Overrides:

operationCount in class AbstractMemtable

partitionCount

public long partitionCount()

Description copied from interface: Memtable

Number of partitions stored in the memtable

getMinTimestamp

public long getMinTimestamp()

Returns the minTS if one available, otherwise NO_MIN_TIMESTAMP. EncodingStats uses a synthetic epoch TS at 2015. We don't want to leak that (CASSANDRA-18118) so we return NO_MIN_TIMESTAMP instead.

Specified by:

getMinTimestamp in interface UnfilteredSource

Overrides:

getMinTimestamp in class AbstractMemtable

Returns:

The minTS or NO_MIN_TIMESTAMP if none available

getMinLocalDeletionTime

public int getMinLocalDeletionTime()

Description copied from interface: UnfilteredSource

Minimum local deletion time in the memtable

Specified by:

getMinLocalDeletionTime in interface UnfilteredSource

Overrides:

getMinLocalDeletionTime in class AbstractMemtable

partitionIterator

public ShardedSkipListMemtable.MemtableUnfilteredPartitionIterator partitionIterator(ColumnFilter columnFilter,
                                                                                     DataRange dataRange,
                                                                                     SSTableReadsListener readsListener)

Description copied from interface: UnfilteredSource

Returns a partition iterator for the given data range.

Parameters:

columnFilter - filter to apply to all returned partitions

dataRange - the partition and clustering range queried

readsListener - a listener used to handle internal read events

rowIterator

public UnfilteredRowIterator rowIterator(DecoratedKey key,
                                         Slices slices,
                                         ColumnFilter selectedColumns,
                                         boolean reversed,
                                         SSTableReadsListener listener)

Description copied from interface: UnfilteredSource

Returns a row iterator for the given partition, applying the specified row and column filters.

Parameters:

key - the partition key

slices - the row ranges to return

selectedColumns - filter to apply to all returned partitions

reversed - true if the content should be returned in reverse order

listener - a listener used to handle internal read events

rowIterator

public UnfilteredRowIterator rowIterator(DecoratedKey key)

getFlushSet

public Memtable.FlushablePartitionSet<AtomicBTreePartition> getFlushSet(PartitionPosition from,
                                                                        PartitionPosition to)

Description copied from interface: Memtable

Get the collection of data between the given partition boundaries in a form suitable for flushing.

factory

public static org.apache.cassandra.db.memtable.ShardedSkipListMemtable.Factory factory(java.util.Map<java.lang.String,java.lang.String> optionsCopy)