Package org.apache.lucene.codecs.lucene99

Class Lucene99HnswScalarQuantizedVectorsFormat

java.lang.Object

org.apache.lucene.codecs.KnnVectorsFormat

org.apache.lucene.codecs.lucene99.Lucene99HnswScalarQuantizedVectorsFormat

All Implemented Interfaces:

NamedSPILoader.NamedSPI

public class Lucene99HnswScalarQuantizedVectorsFormat
extends KnnVectorsFormat

Lucene 9.9 vector format, which encodes numeric vector values into an associated graph connecting the documents having values. The graph is used to power HNSW search. The format consists of two files, and uses Lucene99ScalarQuantizedVectorsFormat to store the actual vectors: For details on graph storage and file extensions, see Lucene99HnswVectorsFormat.

WARNING: This API is experimental and might change in incompatible ways in the next release.

Field Summary

Fields

Modifier and Type	Field	Description
static final String	NAME

Fields inherited from class org.apache.lucene.codecs.KnnVectorsFormat

DEFAULT_MAX_DIMENSIONS, EMPTY

Constructor Summary

Constructors

Constructor	Description
Lucene99HnswScalarQuantizedVectorsFormat()	Constructs a format using default graph construction parameters with 7 bit quantization
Lucene99HnswScalarQuantizedVectorsFormat(int maxConn, int beamWidth)	Constructs a format using the given graph construction parameters with 7 bit quantization
Lucene99HnswScalarQuantizedVectorsFormat(int maxConn, int beamWidth, int numMergeWorkers, int bits, boolean compress, Float confidenceInterval, ExecutorService mergeExec)	Constructs a format using the given graph construction parameters and scalar quantization.

Method Summary

Modifier and Type	Method	Description
KnnVectorsReader	fieldsReader(SegmentReadState state)	Returns a KnnVectorsReader to read the vectors from the index.
KnnVectorsWriter	fieldsWriter(SegmentWriteState state)	Returns a KnnVectorsWriter to write the vectors to the index.
int	getMaxDimensions(String fieldName)	Returns the maximum number of vector dimensions supported by this codec for the given field name
String	toString()

Methods inherited from class org.apache.lucene.codecs.KnnVectorsFormat

availableKnnVectorsFormats, forName, getName, reloadKnnVectorsFormat

Methods inherited from class java.lang.Object

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

Field Details

NAME

public static final String NAME

See Also:

• Constant Field Values

Constructor Details

Lucene99HnswScalarQuantizedVectorsFormat

public Lucene99HnswScalarQuantizedVectorsFormat()

Constructs a format using default graph construction parameters with 7 bit quantization

Lucene99HnswScalarQuantizedVectorsFormat

public Lucene99HnswScalarQuantizedVectorsFormat(int maxConn,
 int beamWidth)

Constructs a format using the given graph construction parameters with 7 bit quantization

Parameters:

maxConn - the maximum number of connections to a node in the HNSW graph

beamWidth - the size of the queue maintained during graph construction.

Lucene99HnswScalarQuantizedVectorsFormat

public Lucene99HnswScalarQuantizedVectorsFormat(int maxConn,
 int beamWidth,
 int numMergeWorkers,
 int bits,
 boolean compress,
 Float confidenceInterval,
 ExecutorService mergeExec)

Constructs a format using the given graph construction parameters and scalar quantization.

Parameters:

maxConn - the maximum number of connections to a node in the HNSW graph

beamWidth - the size of the queue maintained during graph construction.

numMergeWorkers - number of workers (threads) that will be used when doing merge. If larger than 1, a non-null ExecutorService must be passed as mergeExec

bits - the number of bits to use for scalar quantization (must be 4 or 7)

compress - whether to compress the quantized vectors by another 50% when bits=4. If `true`, pairs of (4 bit quantized) dimensions are packed into a single byte. This must be `false` when bits=7. This provides a trade-off of 50% reduction in hot vector memory usage during searching, at some decode speed penalty.

confidenceInterval - the confidenceInterval for scalar quantizing the vectors, when `null` it is calculated based on the vector field dimensions. When `0`, the quantiles are dynamically determined by sampling many confidence intervals and determining the most accurate pair.

mergeExec - the ExecutorService that will be used by ALL vector writers that are generated by this format to do the merge

Method Details

fieldsWriter

public KnnVectorsWriter fieldsWriter(SegmentWriteState state)
                              throws IOException

Description copied from class: KnnVectorsFormat

Returns a KnnVectorsWriter to write the vectors to the index.

Specified by:

fieldsWriter in class KnnVectorsFormat

Throws:

IOException

fieldsReader

public KnnVectorsReader fieldsReader(SegmentReadState state)
                              throws IOException

Description copied from class: KnnVectorsFormat

Returns a KnnVectorsReader to read the vectors from the index.

Specified by:

fieldsReader in class KnnVectorsFormat

Throws:

IOException

getMaxDimensions

public int getMaxDimensions(String fieldName)

Description copied from class: KnnVectorsFormat

Returns the maximum number of vector dimensions supported by this codec for the given field name

Codecs implement this method to specify the maximum number of dimensions they support.

Specified by:

getMaxDimensions in class KnnVectorsFormat

Parameters:

fieldName - the field name

Returns:

the maximum number of vector dimensions.

toString

public String toString()

Overrides:

toString in class Object