Package org.apache.cassandra.db.rows

Class EncodingStats

java.lang.Object

org.apache.cassandra.db.rows.EncodingStats

All Implemented Interfaces:

IMeasurableMemory

public class EncodingStats
extends Object
implements IMeasurableMemory

Stats used for the encoding of the rows and tombstones of a given source.

Those stats are used to optimize the on-wire and on-disk storage of rows. More precisely, the minTimestamp, minLocalDeletionTime and minTTL stats are used to delta-encode those information for the sake of vint encoding.

Note that due to their use, those stats can suffer to be somewhat inaccurate (the more incurrate they are, the less effective the storage will be, but provided the stats are not completly wacky, this shouldn't have too huge an impact on performance) and in fact they will not always be accurate for reasons explained in SerializationHeader.make(org.apache.cassandra.schema.TableMetadata, java.util.Collection<org.apache.cassandra.io.sstable.format.SSTableReader>).

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	EncodingStats.Collector
static class	EncodingStats.Serializer

Field Summary

Fields

Modifier and Type	Field	Description
static final long	HEAP_SIZE
final long	minLocalDeletionTime
final long	minTimestamp
final int	minTTL
static final EncodingStats	NO_STATS
static final EncodingStats.Serializer	serializer
static final long	TIMESTAMP_EPOCH

Constructor Summary

Constructors

Constructor	Description
EncodingStats(long minTimestamp, long minLocalDeletionTime, int minTTL)

Method Summary

Modifier and Type	Method	Description
boolean	equals(Object o)
int	hashCode()
static <V, F extends Function<V, EncodingStats>> EncodingStats	merge(List<V> values, F function)	Merge one or more EncodingStats, that are lazily materialized from some list of arbitrary type by the provided function
EncodingStats	mergeWith(EncodingStats that)	Merge this stats with another one.
String	toString()
long	unsharedHeapSize()

Methods inherited from class java.lang.Object

clone, finalize, getClass, notify, notifyAll, wait, wait, wait

Field Details

TIMESTAMP_EPOCH

public static final long TIMESTAMP_EPOCH

NO_STATS

public static final EncodingStats NO_STATS

HEAP_SIZE

public static final long HEAP_SIZE

serializer

public static final EncodingStats.Serializer serializer

minTimestamp

public final long minTimestamp

minLocalDeletionTime

public final long minLocalDeletionTime

minTTL

public final int minTTL

Constructor Details

EncodingStats

public EncodingStats(long minTimestamp,
 long minLocalDeletionTime,
 int minTTL)

Method Details

mergeWith

public EncodingStats mergeWith(EncodingStats that)

Merge this stats with another one.

The comments of SerializationHeader.make(org.apache.cassandra.schema.TableMetadata, java.util.Collection<org.apache.cassandra.io.sstable.format.SSTableReader>) applies here too, i.e. the result of merging will be not totally accurate but we can live with that.

merge

public static <V,
F extends Function<V,
EncodingStats>>
EncodingStats merge(List<V> values,
 F function)

Merge one or more EncodingStats, that are lazily materialized from some list of arbitrary type by the provided function

equals

public boolean equals(Object o)

Overrides:

equals in class Object

hashCode

public int hashCode()

Overrides:

hashCode in class Object

unsharedHeapSize

public long unsharedHeapSize()

Specified by:

unsharedHeapSize in interface IMeasurableMemory

Returns:

the amount of on-heap memory retained by the object that might be reclaimed if the object were reclaimed, i.e. it should try to exclude globally cached data where possible, or counting portions of arrays that are referenced by the object but used by other objects only (e.g. slabbed byte-buffers), etc.

toString

public String toString()

Overrides:

toString in class Object