org.apache.hadoop.hive.ql.exec.vector
Class VectorAggregationBufferBatch

java.lang.Object
  org.apache.hadoop.hive.ql.exec.vector.VectorAggregationBufferBatch

public class VectorAggregationBufferBatch
extends Object

This maps a batch to the aggregation buffers sets to use for each row (key)

Constructor Summary

VectorAggregationBufferBatch()

Method Summary

void	compileAggregationBatchInfo(VectorAggregateExpression[] aggregators)
VectorAggregationBufferRow[]	getAggregationBuffers() the array of aggregation buffers for the current batch.
int	getAggregatorsFixedSize() Returns the fixed size consumed by the aggregation buffers
int	getDistinctBufferSetCount() number of distinct aggregation buffer sets (ie.
boolean	getHasVariableSize() returns True if any of the aggregators has a variable size
int	getVariableSize(int batchSize)
void	mapAggregationBufferSet(VectorAggregationBufferRow bufferSet, int row) assigns the given aggregation buffer set to a given batch row (by row number).
void	startBatch() resets the internal aggregation buffers sets index and increments the versioning used to optimize the selection vector population.

Methods inherited from class java.lang.Object

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

Constructor Detail

VectorAggregationBufferBatch

public VectorAggregationBufferBatch()

Method Detail

getHasVariableSize

public boolean getHasVariableSize()

returns True if any of the aggregators has a variable size

Returns:

getAggregatorsFixedSize

public int getAggregatorsFixedSize()

Returns the fixed size consumed by the aggregation buffers

Returns:

getAggregationBuffers

public VectorAggregationBufferRow[] getAggregationBuffers()

the array of aggregation buffers for the current batch. content past the getDistinctBufferSetCount() index is stale from previous batches.

Returns:

getDistinctBufferSetCount

public int getDistinctBufferSetCount()

number of distinct aggregation buffer sets (ie. keys) in the current batch.

Returns:

startBatch

public void startBatch()

resets the internal aggregation buffers sets index and increments the versioning used to optimize the selection vector population.

mapAggregationBufferSet

public void mapAggregationBufferSet(VectorAggregationBufferRow bufferSet,
                                    int row)

assigns the given aggregation buffer set to a given batch row (by row number). populates the selection vector appropriately. This is where the versioning numbers play a role in determining if the index cached on the aggregation buffer set is stale.

compileAggregationBatchInfo

public void compileAggregationBatchInfo(VectorAggregateExpression[] aggregators)

getVariableSize

public int getVariableSize(int batchSize)