org.apache.hadoop.hive.ql.udf.generic

Class GenericUDAFComputeStats.GenericUDAFLongStatsEvaluator.LongStatsAgg

java.lang.Object

org.apache.hadoop.hive.ql.udf.generic.GenericUDAFEvaluator.AbstractAggregationBuffer

org.apache.hadoop.hive.ql.udf.generic.GenericUDAFComputeStats.GenericUDAFLongStatsEvaluator.LongStatsAgg

All Implemented Interfaces:

GenericUDAFEvaluator.AggregationBuffer

Enclosing class:

GenericUDAFComputeStats.GenericUDAFLongStatsEvaluator

public static class GenericUDAFComputeStats.GenericUDAFLongStatsEvaluator.LongStatsAgg
extends GenericUDAFEvaluator.AbstractAggregationBuffer

Field Summary

Fields

Modifier and Type	Field and Description
java.lang.String	columnType
long	countNulls
boolean	firstItem
long	max
long	min
int	numBitVectors
LongNumDistinctValueEstimator	numDV

Constructor Summary

Constructors

Constructor and Description
LongStatsAgg()

Method Summary

Modifier and Type	Method and Description
int	estimate() Estimate the size of memory which is occupied by aggregation buffer.

Methods inherited from class java.lang.Object

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

Field Detail

columnType

public java.lang.String columnType

min

public long min

max

public long max

countNulls

public long countNulls

numDV

public LongNumDistinctValueEstimator numDV

firstItem

public boolean firstItem

numBitVectors

public int numBitVectors

Constructor Detail

LongStatsAgg

public LongStatsAgg()

Method Detail

estimate

public int estimate()

Description copied from class: GenericUDAFEvaluator.AbstractAggregationBuffer

Estimate the size of memory which is occupied by aggregation buffer. Currently, hive assumes that primitives types occupies 16 byte and java object has 64 byte overhead for each. For map, each entry also has 64 byte overhead.

Overrides:

estimate in class GenericUDAFEvaluator.AbstractAggregationBuffer