com.hazelcast.cardinality.impl.hyperloglog

Interface HyperLogLog

All Superinterfaces:

DataSerializable, IdentifiedDataSerializable, Versioned

All Known Implementing Classes:

HyperLogLogImpl

public interface HyperLogLog
extends IdentifiedDataSerializable, Versioned

HyperLogLog is a redundant and highly available distributed data-structure used for cardinality estimation purposes on unique items in significantly sized data cultures. HyperLogLog uses P^2 byte registers for storage and computation. HyperLogLog is an implementation of the two famous papers

• http://algo.inria.fr/flajolet/Publications/FlFuGaMe07.pdf
• http://static.googleusercontent.com/media/research.google.com/en//pubs/archive/40671.pdf

Method Summary

Modifier and Type	Method and Description
void	add(long hash) Aggregates the hash in the HyperLogLog registers.
void	addAll(long[] hashes) Batch aggregation of hash values in the HyperLogLog registers.
long	estimate() Computes a new estimate for the current status of the registers.
void	merge(HyperLogLog other) Merge the two HyperLogLog structures in one.

Methods inherited from interface com.hazelcast.nio.serialization.IdentifiedDataSerializable

getFactoryId, getId

Methods inherited from interface com.hazelcast.nio.serialization.DataSerializable

readData, writeData

Method Detail

estimate

long estimate()

Computes a new estimate for the current status of the registers. If it was previously estimated and never invalidated, then a cached version is used.

Returns:

previously cached estimation or a newly computed one.

add

void add(long hash)

Aggregates the hash in the HyperLogLog registers.

Parameters:

hash - the value to aggregate

addAll

void addAll(long[] hashes)

Batch aggregation of hash values in the HyperLogLog registers.

Parameters:

hashes - the hash values array to aggregate

merge

void merge(HyperLogLog other)

Merge the two HyperLogLog structures in one. Estimations from both are taken into consideration and the unified estimate should be similar to the distinct union set of the two.

Parameters:

other - The second HLL to be merged into this one