Package org.apache.druid.query.topn
Class Generic1AggPooledTopNScannerPrototype
java.lang.Object
org.apache.druid.query.topn.Generic1AggPooledTopNScannerPrototype
- All Implemented Interfaces:
Generic1AggPooledTopNScanner
public final class Generic1AggPooledTopNScannerPrototype
extends Object
implements Generic1AggPooledTopNScanner
-
Constructor Summary
Constructors -
Method Summary
Modifier and TypeMethodDescriptionlongscanAndAggregate(DimensionSelector dimensionSelector, BufferAggregator aggregator, int aggregatorSize, Cursor cursor, CursorGranularizer granularizer, int[] positions, ByteBuffer resultsBuffer) Any changes to this method should be coordinated withTopNUtils,PooledTopNAlgorithm.computeSpecializedScanAndAggregateImplementations()and downstream methods.
-
Constructor Details
-
Generic1AggPooledTopNScannerPrototype
public Generic1AggPooledTopNScannerPrototype()
-
-
Method Details
-
scanAndAggregate
public long scanAndAggregate(DimensionSelector dimensionSelector, BufferAggregator aggregator, int aggregatorSize, Cursor cursor, CursorGranularizer granularizer, int[] positions, ByteBuffer resultsBuffer) Any changes to this method should be coordinated withTopNUtils,PooledTopNAlgorithm.computeSpecializedScanAndAggregateImplementations()and downstream methods. It should be checked with a tool like https://github.com/AdoptOpenJDK/jitwatch that C2 compiler output for this method doesn't have any method calls in the while loop, i. e. all method calls are inlined. To be able to see assembly of this method in JITWatch and other similar tools,PooledTopNAlgorithm.SPECIALIZE_GENERIC_ONE_AGG_POOLED_TOPNshould be turned off. Note that in this case the benchmark should be "naturally monomorphic", i. e. execute this method always with the same runtime shape. If the while loop contains not inlined method calls, it should be considered as a performance bug.- Specified by:
scanAndAggregatein interfaceGeneric1AggPooledTopNScanner- Parameters:
aggregatorSize- number of bytes required by aggregator for a single aggregationpositions- a cache for positions in resultsBuffer, where specific (indexed) dimension values are aggregated- Returns:
- number of scanned rows, i. e. number of steps made with the given cursor
-