Skip to content

Apache Lucene open-source search software

Improve int4 compressed comparisons performance (#13321)

Benjamin Trente40e1086f5ab
diff --git a/lucene/CHANGES.txt b/lucene/CHANGES.txt
index 937130f506f..1557689c1e3 100644
--- a/lucene/CHANGES.txt
+++ b/lucene/CHANGES.txt
@@ -325,6 +325,9 @@ Optimizations
* GITHUB#13284: Per-field doc values and knn vectors readers now use a HashMap internally instead of
a TreeMap. (Adrien Grand)
+* GITHUB#13321: Improve compressed int4 quantized vector search by utilizing SIMD inline with the decompression
+ process. (Ben Trent)
+
Bug Fixes
---------------------
diff --git a/lucene/benchmark-jmh/src/java/org/apache/lucene/benchmark/jmh/VectorUtilBenchmark.java b/lucene/benchmark-jmh/src/java/org/apache/lucene/benchmark/jmh/VectorUtilBenchmark.java
index b2c2653baf3..cbef3ecf626 100644
--- a/lucene/benchmark-jmh/src/java/org/apache/lucene/benchmark/jmh/VectorUtilBenchmark.java
+++ b/lucene/benchmark-jmh/src/java/org/apache/lucene/benchmark/jmh/VectorUtilBenchmark.java
@@ -33,13 +33,21 @@ import org.openjdk.jmh.annotations.*;
value = 3,
jvmArgsAppend = {"-Xmx2g", "-Xms2g", "-XX:+AlwaysPreTouch"})
public class VectorUtilBenchmark {
+ static void compressBytes(byte[] raw, byte[] compressed) {
+ for (int i = 0; i < compressed.length; ++i) {
+ int v = (raw[i] << 4) | raw[compressed.length + i];
+ compressed[i] = (byte) v;
+ }
+ }
private byte[] bytesA;
private byte[] bytesB;
private byte[] halfBytesA;
private byte[] halfBytesB;
+ private byte[] halfBytesBPacked;
private float[] floatsA;
private float[] floatsB;
+ private int expectedhalfByteDotProduct;
@Param({"1", "128", "207", "256", "300", "512", "702", "1024"})
int size;
@@ -55,11 +63,18 @@ public class VectorUtilBenchmark {
random.nextBytes(bytesB);
// random half byte arrays for binary methods
// this means that all values must be between 0 and 15
+ expectedhalfByteDotProduct = 0;
halfBytesA = new byte[size];
halfBytesB = new byte[size];
for (int i = 0; i < size; ++i) {
halfBytesA[i] = (byte) random.nextInt(16);
halfBytesB[i] = (byte) random.nextInt(16);
+ expectedhalfByteDotProduct += halfBytesA[i] * halfBytesB[i];
+ }
+ // pack the half byte arrays
+ if (size % 2 == 0) {
+ halfBytesBPacked = new byte[(size + 1) >> 1];
+ compressBytes(halfBytesB, halfBytesBPacked);
}
// random float arrays for float methods
@@ -115,6 +130,31 @@ public class VectorUtilBenchmark {
return VectorUtil.int4DotProduct(halfBytesA, halfBytesB);
}
+ @Benchmark
+ public int binaryHalfByteScalarPacked() {
+ if (size % 2 != 0) {
+ throw new RuntimeException("Size must be even for this benchmark");
+ }
+ int v = VectorUtil.int4DotProductPacked(halfBytesA, halfBytesBPacked);
+ if (v != expectedhalfByteDotProduct) {
+ throw new RuntimeException("Expected " + expectedhalfByteDotProduct + " but got " + v);
+ }
+ return v;
+ }
+
+ @Benchmark
+ @Fork(jvmArgsPrepend = {"--add-modules=jdk.incubator.vector"})
+ public int binaryHalfByteVectorPacked() {
+ if (size % 2 != 0) {
+ throw new RuntimeException("Size must be even for this benchmark");
+ }
+ int v = VectorUtil.int4DotProductPacked(halfBytesA, halfBytesBPacked);
+ if (v != expectedhalfByteDotProduct) {
+ throw new RuntimeException("Expected " + expectedhalfByteDotProduct + " but got " + v);
+ }
+ return v;
+ }
+
@Benchmark
public float floatCosineScalar() {
return VectorUtil.cosine(floatsA, floatsB);
diff --git a/lucene/core/src/java/org/apache/lucene/codecs/hnsw/ScalarQuantizedVectorScorer.java b/lucene/core/src/java/org/apache/lucene/codecs/hnsw/ScalarQuantizedVectorScorer.java
index 47dc18afc0a..9914d2cb537 100644
--- a/lucene/core/src/java/org/apache/lucene/codecs/hnsw/ScalarQuantizedVectorScorer.java
+++ b/lucene/core/src/java/org/apache/lucene/codecs/hnsw/ScalarQuantizedVectorScorer.java
@@ -19,12 +19,12 @@ package org.apache.lucene.codecs.hnsw;
import java.io.IOException;
import org.apache.lucene.index.VectorSimilarityFunction;
+import org.apache.lucene.util.ArrayUtil;
+import org.apache.lucene.util.VectorUtil;
import org.apache.lucene.util.hnsw.RandomAccessVectorValues;
import org.apache.lucene.util.hnsw.RandomVectorScorer;
import org.apache.lucene.util.hnsw.RandomVectorScorerSupplier;
import org.apache.lucene.util.quantization.RandomAccessQuantizedByteVectorValues;
-import org.apache.lucene.util.quantization.ScalarQuantizedRandomVectorScorer;
-import org.apache.lucene.util.quantization.ScalarQuantizedRandomVectorScorerSupplier;
import org.apache.lucene.util.quantization.ScalarQuantizedVectorSimilarity;
import org.apache.lucene.util.quantization.ScalarQuantizer;
@@ -35,6 +35,23 @@ import org.apache.lucene.util.quantization.ScalarQuantizer;
*/
public class ScalarQuantizedVectorScorer implements FlatVectorsScorer {
+ public static float quantizeQuery(
+ float[] query,
+ byte[] quantizedQuery,
+ VectorSimilarityFunction similarityFunction,
+ ScalarQuantizer scalarQuantizer) {
+ float[] processedQuery =
+ switch (similarityFunction) {
+ case EUCLIDEAN, DOT_PRODUCT, MAXIMUM_INNER_PRODUCT -> query;
+ case COSINE -> {
+ float[] queryCopy = ArrayUtil.copyOfSubArray(query, 0, query.length);
+ VectorUtil.l2normalize(queryCopy);
+ yield queryCopy;
+ }
+ };
+ return scalarQuantizer.quantize(processedQuery, quantizedQuery, similarityFunction);
+ }
+
private final FlatVectorsScorer nonQuantizedDelegate;
public ScalarQuantizedVectorScorer(FlatVectorsScorer flatVectorsScorer) {
@@ -65,18 +82,21 @@ public class ScalarQuantizedVectorScorer implements FlatVectorsScorer {
ScalarQuantizer scalarQuantizer = quantizedByteVectorValues.getScalarQuantizer();
byte[] targetBytes = new byte[target.length];
float offsetCorrection =
- ScalarQuantizedRandomVectorScorer.quantizeQuery(
- target, targetBytes, similarityFunction, scalarQuantizer);
+ quantizeQuery(target, targetBytes, similarityFunction, scalarQuantizer);
ScalarQuantizedVectorSimilarity scalarQuantizedVectorSimilarity =
ScalarQuantizedVectorSimilarity.fromVectorSimilarity(
similarityFunction,
scalarQuantizer.getConstantMultiplier(),
scalarQuantizer.getBits());
- return new ScalarQuantizedRandomVectorScorer(
- scalarQuantizedVectorSimilarity,
- quantizedByteVectorValues,
- targetBytes,
- offsetCorrection);
+ return new RandomVectorScorer.AbstractRandomVectorScorer(quantizedByteVectorValues) {
+ @Override
+ public float score(int node) throws IOException {
+ byte[] nodeVector = quantizedByteVectorValues.vectorValue(node);
+ float nodeOffset = quantizedByteVectorValues.getScoreCorrectionConstant(node);
+ return scalarQuantizedVectorSimilarity.score(
+ targetBytes, offsetCorrection, nodeVector, nodeOffset);
+ }
+ };
}
// It is possible to get to this branch during initial indexing and flush
return nonQuantizedDelegate.getRandomVectorScorer(similarityFunction, vectorValues, target);
@@ -95,4 +115,60 @@ public class ScalarQuantizedVectorScorer implements FlatVectorsScorer {
public String toString() {
return "ScalarQuantizedVectorScorer(" + "nonQuantizedDelegate=" + nonQuantizedDelegate + ')';
}
+
+ /**
+ * Quantized vector scorer supplier
+ *
+ * @lucene.experimental
+ */
+ public static class ScalarQuantizedRandomVectorScorerSupplier
+ implements RandomVectorScorerSupplier {
+
+ private final RandomAccessQuantizedByteVectorValues values;
+ private final ScalarQuantizedVectorSimilarity similarity;
+ private final VectorSimilarityFunction vectorSimilarityFunction;
+
+ public ScalarQuantizedRandomVectorScorerSupplier(
+ VectorSimilarityFunction similarityFunction,
+ ScalarQuantizer scalarQuantizer,
+ RandomAccessQuantizedByteVectorValues values) {
+ this.similarity =
+ ScalarQuantizedVectorSimilarity.fromVectorSimilarity(
+ similarityFunction,
+ scalarQuantizer.getConstantMultiplier(),
+ scalarQuantizer.getBits());
+ this.values = values;
+ this.vectorSimilarityFunction = similarityFunction;
+ }
+
+ private ScalarQuantizedRandomVectorScorerSupplier(
+ ScalarQuantizedVectorSimilarity similarity,
+ VectorSimilarityFunction vectorSimilarityFunction,
+ RandomAccessQuantizedByteVectorValues values) {
+ this.similarity = similarity;
+ this.values = values;
+ this.vectorSimilarityFunction = vectorSimilarityFunction;
+ }
+
+ @Override
+ public RandomVectorScorer scorer(int ord) throws IOException {
+ final RandomAccessQuantizedByteVectorValues vectorsCopy = values.copy();
+ final byte[] queryVector = values.vectorValue(ord);
+ final float queryOffset = values.getScoreCorrectionConstant(ord);
+ return new RandomVectorScorer.AbstractRandomVectorScorer(vectorsCopy) {
+ @Override
+ public float score(int node) throws IOException {
+ byte[] nodeVector = vectorsCopy.vectorValue(node);
+ float nodeOffset = vectorsCopy.getScoreCorrectionConstant(node);
+ return similarity.score(queryVector, queryOffset, nodeVector, nodeOffset);
+ }
+ };
+ }
+
+ @Override
+ public RandomVectorScorerSupplier copy() throws IOException {
+ return new ScalarQuantizedRandomVectorScorerSupplier(
+ similarity, vectorSimilarityFunction, values.copy());
+ }
+ }
}
diff --git a/lucene/core/src/java/org/apache/lucene/codecs/lucene95/OffHeapByteVectorValues.java b/lucene/core/src/java/org/apache/lucene/codecs/lucene95/OffHeapByteVectorValues.java
index da11df1e251..8d98a9cd1c8 100644
--- a/lucene/core/src/java/org/apache/lucene/codecs/lucene95/OffHeapByteVectorValues.java
+++ b/lucene/core/src/java/org/apache/lucene/codecs/lucene95/OffHeapByteVectorValues.java
@@ -68,6 +68,11 @@ public abstract class OffHeapByteVectorValues extends ByteVectorValues
return binaryValue;
}
+ @Override
+ public IndexInput getSlice() {
+ return slice;
+ }
+
private void readValue(int targetOrd) throws IOException {
slice.seek((long) targetOrd * byteSize);
slice.readBytes(byteBuffer.array(), byteBuffer.arrayOffset(), byteSize);
diff --git a/lucene/core/src/java/org/apache/lucene/codecs/lucene95/OffHeapFloatVectorValues.java b/lucene/core/src/java/org/apache/lucene/codecs/lucene95/OffHeapFloatVectorValues.java
index a13c9f55bb1..0aeddaf1536 100644
--- a/lucene/core/src/java/org/apache/lucene/codecs/lucene95/OffHeapFloatVectorValues.java
+++ b/lucene/core/src/java/org/apache/lucene/codecs/lucene95/OffHeapFloatVectorValues.java
@@ -56,6 +56,11 @@ public abstract class OffHeapFloatVectorValues extends FloatVectorValues
return size;
}
+ @Override
+ public IndexInput getSlice() {
+ return slice;
+ }
+
@Override
public float[] vectorValue(int targetOrd) throws IOException {
if (lastOrd == targetOrd) {
diff --git a/lucene/core/src/java/org/apache/lucene/codecs/lucene99/Lucene99ScalarQuantizedVectorScorer.java b/lucene/core/src/java/org/apache/lucene/codecs/lucene99/Lucene99ScalarQuantizedVectorScorer.java
new file mode 100644
index 00000000000..edb35595ec1
--- /dev/null
+++ b/lucene/core/src/java/org/apache/lucene/codecs/lucene99/Lucene99ScalarQuantizedVectorScorer.java
@@ -0,0 +1,295 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.lucene.codecs.lucene99;
+
+import static org.apache.lucene.codecs.hnsw.ScalarQuantizedVectorScorer.quantizeQuery;
+
+import java.io.IOException;
+import org.apache.lucene.codecs.hnsw.FlatVectorsScorer;
+import org.apache.lucene.index.VectorSimilarityFunction;
+import org.apache.lucene.util.VectorUtil;
+import org.apache.lucene.util.hnsw.RandomAccessVectorValues;
+import org.apache.lucene.util.hnsw.RandomVectorScorer;
+import org.apache.lucene.util.hnsw.RandomVectorScorerSupplier;
+import org.apache.lucene.util.quantization.RandomAccessQuantizedByteVectorValues;
+import org.apache.lucene.util.quantization.ScalarQuantizer;
+
+/**
+ * Optimized scalar quantized implementation of {@link FlatVectorsScorer} for quantized vectors
+ * stored in the Lucene99 format.
+ *
+ * @lucene.experimental
+ */
+public class Lucene99ScalarQuantizedVectorScorer implements FlatVectorsScorer {
+
+ private final FlatVectorsScorer nonQuantizedDelegate;
+
+ public Lucene99ScalarQuantizedVectorScorer(FlatVectorsScorer flatVectorsScorer) {
+ nonQuantizedDelegate = flatVectorsScorer;
+ }
+
+ @Override
+ public RandomVectorScorerSupplier getRandomVectorScorerSupplier(
+ VectorSimilarityFunction similarityFunction, RandomAccessVectorValues vectorValues)
+ throws IOException {
+ if (vectorValues instanceof RandomAccessQuantizedByteVectorValues quantizedByteVectorValues) {
+ return new ScalarQuantizedRandomVectorScorerSupplier(
+ quantizedByteVectorValues, similarityFunction);
+ }
+ // It is possible to get to this branch during initial indexing and flush
+ return nonQuantizedDelegate.getRandomVectorScorerSupplier(similarityFunction, vectorValues);
+ }
+
+ @Override
+ public RandomVectorScorer getRandomVectorScorer(
+ VectorSimilarityFunction similarityFunction,
+ RandomAccessVectorValues vectorValues,
+ float[] target)
+ throws IOException {
+ if (vectorValues instanceof RandomAccessQuantizedByteVectorValues quantizedByteVectorValues) {
+ ScalarQuantizer scalarQuantizer = quantizedByteVectorValues.getScalarQuantizer();
+ byte[] targetBytes = new byte[target.length];
+ float offsetCorrection =
+ quantizeQuery(target, targetBytes, similarityFunction, scalarQuantizer);
+ return fromVectorSimilarity(
+ targetBytes,
+ offsetCorrection,
+ similarityFunction,
+ scalarQuantizer.getConstantMultiplier(),
+ quantizedByteVectorValues);
+ }
+ // It is possible to get to this branch during initial indexing and flush
+ return nonQuantizedDelegate.getRandomVectorScorer(similarityFunction, vectorValues, target);
+ }
+
+ @Override
+ public RandomVectorScorer getRandomVectorScorer(
+ VectorSimilarityFunction similarityFunction,
+ RandomAccessVectorValues vectorValues,
+ byte[] target)
+ throws IOException {
+ return nonQuantizedDelegate.getRandomVectorScorer(similarityFunction, vectorValues, target);
+ }
+
+ @Override
+ public String toString() {
+ return "ScalarQuantizedVectorScorer(" + "nonQuantizedDelegate=" + nonQuantizedDelegate + ')';
+ }
+
+ static RandomVectorScorer fromVectorSimilarity(
+ byte[] targetBytes,
+ float offsetCorrection,
+ VectorSimilarityFunction sim,
+ float constMultiplier,
+ RandomAccessQuantizedByteVectorValues values) {
+ return switch (sim) {
+ case EUCLIDEAN -> new Euclidean(values, constMultiplier, targetBytes);
+ case COSINE, DOT_PRODUCT -> dotProductFactory(
+ targetBytes, offsetCorrection, sim, constMultiplier, values, f -> (1 + f) / 2);
+ case MAXIMUM_INNER_PRODUCT -> dotProductFactory(
+ targetBytes,
+ offsetCorrection,
+ sim,
+ constMultiplier,
+ values,
+ VectorUtil::scaleMaxInnerProductScore);
+ };
+ }
+
+ private static RandomVectorScorer.AbstractRandomVectorScorer dotProductFactory(
+ byte[] targetBytes,
+ float offsetCorrection,
+ VectorSimilarityFunction sim,
+ float constMultiplier,
+ RandomAccessQuantizedByteVectorValues values,
+ FloatToFloatFunction scoreAdjustmentFunction) {
+ if (values.getScalarQuantizer().getBits() <= 4) {
+ if (values.getVectorByteLength() != values.dimension() && values.getSlice() != null) {
+ return new CompressedInt4DotProduct(
+ values, constMultiplier, targetBytes, offsetCorrection, scoreAdjustmentFunction);
+ }
+ return new Int4DotProduct(
+ values, constMultiplier, targetBytes, offsetCorrection, scoreAdjustmentFunction);
+ }
+ return new DotProduct(
+ values, constMultiplier, targetBytes, offsetCorrection, scoreAdjustmentFunction);
+ }
+
+ private static class Euclidean extends RandomVectorScorer.AbstractRandomVectorScorer {
+ private final float constMultiplier;
+ private final byte[] targetBytes;
+ private final RandomAccessQuantizedByteVectorValues values;
+
+ private Euclidean(
+ RandomAccessQuantizedByteVectorValues values, float constMultiplier, byte[] targetBytes) {
+ super(values);
+ this.values = values;
+ this.constMultiplier = constMultiplier;
+ this.targetBytes = targetBytes;
+ }
+
+ @Override
+ public float score(int node) throws IOException {
+ byte[] nodeVector = values.vectorValue(node);
+ int squareDistance = VectorUtil.squareDistance(nodeVector, targetBytes);
+ float adjustedDistance = squareDistance * constMultiplier;
+ return 1 / (1f + adjustedDistance);
+ }
+ }
+
+ /** Calculates dot product on quantized vectors, applying the appropriate corrections */
+ private static class DotProduct extends RandomVectorScorer.AbstractRandomVectorScorer {
+ private final float constMultiplier;
+ private final RandomAccessQuantizedByteVectorValues values;
+ private final byte[] targetBytes;
+ private final float offsetCorrection;
+ private final FloatToFloatFunction scoreAdjustmentFunction;
+
+ public DotProduct(
+ RandomAccessQuantizedByteVectorValues values,
+ float constMultiplier,
+ byte[] targetBytes,
+ float offsetCorrection,
+ FloatToFloatFunction scoreAdjustmentFunction) {
+ super(values);
+ this.constMultiplier = constMultiplier;
+ this.values = values;
+ this.targetBytes = targetBytes;
+ this.offsetCorrection = offsetCorrection;
+ this.scoreAdjustmentFunction = scoreAdjustmentFunction;
+ }
+
+ @Override
+ public float score(int vectorOrdinal) throws IOException {
+ byte[] storedVector = values.vectorValue(vectorOrdinal);
+ float vectorOffset = values.getScoreCorrectionConstant(vectorOrdinal);
+ int dotProduct = VectorUtil.dotProduct(storedVector, targetBytes);
+ float adjustedDistance = dotProduct * constMultiplier + offsetCorrection + vectorOffset;
+ return scoreAdjustmentFunction.apply(adjustedDistance);
+ }
+ }
+
+ private static class CompressedInt4DotProduct
+ extends RandomVectorScorer.AbstractRandomVectorScorer {
+ private final float constMultiplier;
+ private final RandomAccessQuantizedByteVectorValues values;
+ private final byte[] compressedVector;
+ private final byte[] targetBytes;
+ private final float offsetCorrection;
+ private final FloatToFloatFunction scoreAdjustmentFunction;
+
+ private CompressedInt4DotProduct(
+ RandomAccessQuantizedByteVectorValues values,
+ float constMultiplier,
+ byte[] targetBytes,
+ float offsetCorrection,
+ FloatToFloatFunction scoreAdjustmentFunction) {
+ super(values);
+ this.constMultiplier = constMultiplier;
+ this.values = values;
+ this.compressedVector = new byte[values.getVectorByteLength()];
+ this.targetBytes = targetBytes;
+ this.offsetCorrection = offsetCorrection;
+ this.scoreAdjustmentFunction = scoreAdjustmentFunction;
+ }
+
+ @Override
+ public float score(int vectorOrdinal) throws IOException {
+ // get compressed vector, in Lucene99, vector values are stored and have a single value for
+ // offset correction
+ values.getSlice().seek((long) vectorOrdinal * (values.getVectorByteLength() + Float.BYTES));
+ values.getSlice().readBytes(compressedVector, 0, compressedVector.length);
+ float vectorOffset = values.getScoreCorrectionConstant(vectorOrdinal);
+ int dotProduct = VectorUtil.int4DotProductPacked(targetBytes, compressedVector);
+ float adjustedDistance = dotProduct * constMultiplier + offsetCorrection + vectorOffset;
+ return scoreAdjustmentFunction.apply(adjustedDistance);
+ }
+ }
+
+ private static class Int4DotProduct extends RandomVectorScorer.AbstractRandomVectorScorer {
+ private final float constMultiplier;
+ private final RandomAccessQuantizedByteVectorValues values;
+ private final byte[] targetBytes;
+ private final float offsetCorrection;
+ private final FloatToFloatFunction scoreAdjustmentFunction;
+
+ public Int4DotProduct(
+ RandomAccessQuantizedByteVectorValues values,
+ float constMultiplier,
+ byte[] targetBytes,
+ float offsetCorrection,
+ FloatToFloatFunction scoreAdjustmentFunction) {
+ super(values);
+ this.constMultiplier = constMultiplier;
+ this.values = values;
+ this.targetBytes = targetBytes;
+ this.offsetCorrection = offsetCorrection;
+ this.scoreAdjustmentFunction = scoreAdjustmentFunction;
+ }
+
+ @Override
+ public float score(int vectorOrdinal) throws IOException {
+ byte[] storedVector = values.vectorValue(vectorOrdinal);
+ float vectorOffset = values.getScoreCorrectionConstant(vectorOrdinal);
+ int dotProduct = VectorUtil.int4DotProduct(storedVector, targetBytes);
+ float adjustedDistance = dotProduct * constMultiplier + offsetCorrection + vectorOffset;
+ return scoreAdjustmentFunction.apply(adjustedDistance);
+ }
+ }
+
+ @FunctionalInterface
+ private interface FloatToFloatFunction {
+ float apply(float f);
+ }
+
+ private static final class ScalarQuantizedRandomVectorScorerSupplier
+ implements RandomVectorScorerSupplier {
+
+ private final VectorSimilarityFunction vectorSimilarityFunction;
+ private final RandomAccessQuantizedByteVectorValues values;
+ private final RandomAccessQuantizedByteVectorValues values1;
+ private final RandomAccessQuantizedByteVectorValues values2;
+
+ public ScalarQuantizedRandomVectorScorerSupplier(
+ RandomAccessQuantizedByteVectorValues values,
+ VectorSimilarityFunction vectorSimilarityFunction)
+ throws IOException {
+ this.values = values;
+ this.values1 = values.copy();
+ this.values2 = values.copy();
+ this.vectorSimilarityFunction = vectorSimilarityFunction;
+ }
+
+ @Override
+ public RandomVectorScorer scorer(int ord) throws IOException {
+ byte[] vectorValue = values1.vectorValue(ord);
+ float offsetCorrection = values1.getScoreCorrectionConstant(ord);
+ return fromVectorSimilarity(
+ vectorValue,
+ offsetCorrection,
+ vectorSimilarityFunction,
+ values.getScalarQuantizer().getConstantMultiplier(),
+ values2);
+ }
+
+ @Override
+ public ScalarQuantizedRandomVectorScorerSupplier copy() throws IOException {
+ return new ScalarQuantizedRandomVectorScorerSupplier(values.copy(), vectorSimilarityFunction);
+ }
+ }
+}
diff --git a/lucene/core/src/java/org/apache/lucene/codecs/lucene99/Lucene99ScalarQuantizedVectorsFormat.java b/lucene/core/src/java/org/apache/lucene/codecs/lucene99/Lucene99ScalarQuantizedVectorsFormat.java
index a3d894e64e9..c10f87da2a6 100644
--- a/lucene/core/src/java/org/apache/lucene/codecs/lucene99/Lucene99ScalarQuantizedVectorsFormat.java
+++ b/lucene/core/src/java/org/apache/lucene/codecs/lucene99/Lucene99ScalarQuantizedVectorsFormat.java
@@ -22,7 +22,6 @@ import org.apache.lucene.codecs.hnsw.DefaultFlatVectorScorer;
import org.apache.lucene.codecs.hnsw.FlatVectorsFormat;
import org.apache.lucene.codecs.hnsw.FlatVectorsReader;
import org.apache.lucene.codecs.hnsw.FlatVectorsWriter;
-import org.apache.lucene.codecs.hnsw.ScalarQuantizedVectorScorer;
import org.apache.lucene.index.SegmentReadState;
import org.apache.lucene.index.SegmentWriteState;
@@ -65,7 +64,7 @@ public class Lucene99ScalarQuantizedVectorsFormat extends FlatVectorsFormat {
final byte bits;
final boolean compress;
- final ScalarQuantizedVectorScorer flatVectorScorer;
+ final Lucene99ScalarQuantizedVectorScorer flatVectorScorer;
/** Constructs a format using default graph construction parameters */
public Lucene99ScalarQuantizedVectorsFormat() {
@@ -102,7 +101,7 @@ public class Lucene99ScalarQuantizedVectorsFormat extends FlatVectorsFormat {
this.bits = (byte) bits;
this.confidenceInterval = confidenceInterval;
this.compress = compress;
- this.flatVectorScorer = new ScalarQuantizedVectorScorer(new DefaultFlatVectorScorer());
+ this.flatVectorScorer = new Lucene99ScalarQuantizedVectorScorer(new DefaultFlatVectorScorer());
}
public static float calculateDefaultConfidenceInterval(int vectorDimension) {
diff --git a/lucene/core/src/java/org/apache/lucene/codecs/lucene99/OffHeapQuantizedByteVectorValues.java b/lucene/core/src/java/org/apache/lucene/codecs/lucene99/OffHeapQuantizedByteVectorValues.java
index 08e666d51ff..9659eb13187 100644
--- a/lucene/core/src/java/org/apache/lucene/codecs/lucene99/OffHeapQuantizedByteVectorValues.java
+++ b/lucene/core/src/java/org/apache/lucene/codecs/lucene99/OffHeapQuantizedByteVectorValues.java
@@ -135,6 +135,26 @@ public abstract class OffHeapQuantizedByteVectorValues extends QuantizedByteVect
return scoreCorrectionConstant[0];
}
+ @Override
+ public float getScoreCorrectionConstant(int targetOrd) throws IOException {
+ if (lastOrd == targetOrd) {
+ return scoreCorrectionConstant[0];
+ }
+ slice.seek(((long) targetOrd * byteSize) + numBytes);
+ slice.readFloats(scoreCorrectionConstant, 0, 1);
+ return scoreCorrectionConstant[0];
+ }
+
+ @Override
+ public IndexInput getSlice() {
+ return slice;
+ }
+
+ @Override
+ public int getVectorByteLength() {
+ return numBytes;
+ }
+
public static OffHeapQuantizedByteVectorValues load(
OrdToDocDISIReaderConfiguration configuration,
int dimension,
diff --git a/lucene/core/src/java/org/apache/lucene/internal/vectorization/DefaultVectorUtilSupport.java b/lucene/core/src/java/org/apache/lucene/internal/vectorization/DefaultVectorUtilSupport.java
index e56d6b97f31..eb5160a0f0d 100644
--- a/lucene/core/src/java/org/apache/lucene/internal/vectorization/DefaultVectorUtilSupport.java
+++ b/lucene/core/src/java/org/apache/lucene/internal/vectorization/DefaultVectorUtilSupport.java
@@ -152,7 +152,21 @@ final class DefaultVectorUtilSupport implements VectorUtilSupport {
}
@Override
- public int int4DotProduct(byte[] a, byte[] b) {
+ public int int4DotProduct(byte[] a, boolean apacked, byte[] b, boolean bpacked) {
+ assert (apacked && bpacked) == false;
+ if (apacked || bpacked) {
+ byte[] packed = apacked ? a : b;
+ byte[] unpacked = apacked ? b : a;
+ int total = 0;
+ for (int i = 0; i < packed.length; i++) {
+ byte packedByte = packed[i];
+ byte unpacked1 = unpacked[i];
+ byte unpacked2 = unpacked[i + packed.length];
+ total += (packedByte & 0x0F) * unpacked2;
+ total += ((packedByte & 0xFF) >> 4) * unpacked1;
+ }
+ return total;
+ }
return dotProduct(a, b);
}
diff --git a/lucene/core/src/java/org/apache/lucene/internal/vectorization/VectorUtilSupport.java b/lucene/core/src/java/org/apache/lucene/internal/vectorization/VectorUtilSupport.java
index 246cbdf95bc..22e5e96aa25 100644
--- a/lucene/core/src/java/org/apache/lucene/internal/vectorization/VectorUtilSupport.java
+++ b/lucene/core/src/java/org/apache/lucene/internal/vectorization/VectorUtilSupport.java
@@ -37,7 +37,7 @@ public interface VectorUtilSupport {
int dotProduct(byte[] a, byte[] b);
/** Returns the dot product over the computed bytes, assuming the values are int4 encoded. */
- int int4DotProduct(byte[] a, byte[] b);
+ int int4DotProduct(byte[] a, boolean apacked, byte[] b, boolean bpacked);
/** Returns the cosine similarity between the two byte vectors. */
float cosine(byte[] a, byte[] b);
diff --git a/lucene/core/src/java/org/apache/lucene/util/VectorUtil.java b/lucene/core/src/java/org/apache/lucene/util/VectorUtil.java
index 7409a1de4cf..a9acca64d69 100644
--- a/lucene/core/src/java/org/apache/lucene/util/VectorUtil.java
+++ b/lucene/core/src/java/org/apache/lucene/util/VectorUtil.java
@@ -179,7 +179,30 @@ public final class VectorUtil {
if (a.length != b.length) {
throw new IllegalArgumentException("vector dimensions differ: " + a.length + "!=" + b.length);
}
- return IMPL.int4DotProduct(a, b);
+ return IMPL.int4DotProduct(a, false, b, false);
+ }
+
+ /**
+ * Dot product computed over int4 (values between [0,15]) bytes. The second vector is considered
+ * "packed" (i.e. every byte representing two values). The following packing is assumed:
+ *
+ * <pre class="prettyprint lang-java">
+ * packed[0] = (raw[0] * 16) | raw[packed.length];
+ * packed[1] = (raw[1] * 16) | raw[packed.length + 1];
+ * ...
+ * packed[packed.length - 1] = (raw[packed.length - 1] * 16) | raw[2 * packed.length - 1];
+ * </pre>
+ *
+ * @param unpacked the unpacked vector, of even length
+ * @param packed the packed vector, of length {@code (unpacked.length + 1) / 2}
+ * @return the value of the dot product of the two vectors
+ */
+ public static int int4DotProductPacked(byte[] unpacked, byte[] packed) {
+ if (packed.length != ((unpacked.length + 1) >> 1)) {
+ throw new IllegalArgumentException(
+ "vector dimensions differ: " + unpacked.length + "!= 2 * " + packed.length);
+ }
+ return IMPL.int4DotProduct(unpacked, false, packed, true);
}
/**
diff --git a/lucene/core/src/java/org/apache/lucene/util/hnsw/RandomAccessVectorValues.java b/lucene/core/src/java/org/apache/lucene/util/hnsw/RandomAccessVectorValues.java
index ecf5339cd21..e2c7372b667 100644
--- a/lucene/core/src/java/org/apache/lucene/util/hnsw/RandomAccessVectorValues.java
+++ b/lucene/core/src/java/org/apache/lucene/util/hnsw/RandomAccessVectorValues.java
@@ -19,6 +19,7 @@ package org.apache.lucene.util.hnsw;
import java.io.IOException;
import java.util.List;
+import org.apache.lucene.store.IndexInput;
import org.apache.lucene.util.Bits;
/**
@@ -41,6 +42,17 @@ public interface RandomAccessVectorValues {
*/
RandomAccessVectorValues copy() throws IOException;
+ /**
+ * Returns a slice of the underlying {@link IndexInput} that contains the vector values if
+ * available
+ */
+ default IndexInput getSlice() {
+ return null;
+ }
+
+ /** Returns the byte length of the vector values. */
+ int getVectorByteLength();
+
/**
* Translates vector ordinal to the correct document ID. By default, this is an identity function.
*
@@ -72,6 +84,12 @@ public interface RandomAccessVectorValues {
* @param targetOrd a valid ordinal, &ge; 0 and &lt; {@link #size()}.
*/
float[] vectorValue(int targetOrd) throws IOException;
+
+ /** Returns the vector byte length, defaults to dimension multiplied by float byte size */
+ @Override
+ default int getVectorByteLength() {
+ return dimension() * Float.BYTES;
+ }
}
/** Byte vector values. */
@@ -85,6 +103,12 @@ public interface RandomAccessVectorValues {
* @param targetOrd a valid ordinal, &ge; 0 and &lt; {@link #size()}.
*/
byte[] vectorValue(int targetOrd) throws IOException;
+
+ /** Returns the vector byte length, defaults to dimension multiplied by byte size */
+ @Override
+ default int getVectorByteLength() {
+ return dimension() * Byte.BYTES;
+ }
}
/**
@@ -107,12 +131,12 @@ public interface RandomAccessVectorValues {
}
@Override
- public float[] vectorValue(int targetOrd) throws IOException {
+ public float[] vectorValue(int targetOrd) {
return vectors.get(targetOrd);
}
@Override
- public RandomAccessVectorValues.Floats copy() throws IOException {
+ public RandomAccessVectorValues.Floats copy() {
return this;
}
};
@@ -138,12 +162,12 @@ public interface RandomAccessVectorValues {
}
@Override
- public byte[] vectorValue(int targetOrd) throws IOException {
+ public byte[] vectorValue(int targetOrd) {
return vectors.get(targetOrd);
}
@Override
- public RandomAccessVectorValues.Bytes copy() throws IOException {
+ public RandomAccessVectorValues.Bytes copy() {
return this;
}
};
diff --git a/lucene/core/src/java/org/apache/lucene/util/quantization/RandomAccessQuantizedByteVectorValues.java b/lucene/core/src/java/org/apache/lucene/util/quantization/RandomAccessQuantizedByteVectorValues.java
index 08b0b6e5a7a..b86009a690e 100644
--- a/lucene/core/src/java/org/apache/lucene/util/quantization/RandomAccessQuantizedByteVectorValues.java
+++ b/lucene/core/src/java/org/apache/lucene/util/quantization/RandomAccessQuantizedByteVectorValues.java
@@ -29,7 +29,7 @@ public interface RandomAccessQuantizedByteVectorValues extends RandomAccessVecto
ScalarQuantizer getScalarQuantizer();
- float getScoreCorrectionConstant();
+ float getScoreCorrectionConstant(int vectorOrd) throws IOException;
@Override
RandomAccessQuantizedByteVectorValues copy() throws IOException;
diff --git a/lucene/core/src/java/org/apache/lucene/util/quantization/ScalarQuantizedRandomVectorScorer.java b/lucene/core/src/java/org/apache/lucene/util/quantization/ScalarQuantizedRandomVectorScorer.java
deleted file mode 100644
index 41fcc3f97d0..00000000000
--- a/lucene/core/src/java/org/apache/lucene/util/quantization/ScalarQuantizedRandomVectorScorer.java
+++ /dev/null
@@ -1,74 +0,0 @@
-/*
- * Licensed to the Apache Software Foundation (ASF) under one or more
- * contributor license agreements. See the NOTICE file distributed with
- * this work for additional information regarding copyright ownership.
- * The ASF licenses this file to You under the Apache License, Version 2.0
- * (the "License"); you may not use this file except in compliance with
- * the License. You may obtain a copy of the License at
- *
- * http://www.apache.org/licenses/LICENSE-2.0
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-package org.apache.lucene.util.quantization;
-
-import java.io.IOException;
-import org.apache.lucene.index.VectorSimilarityFunction;
-import org.apache.lucene.util.ArrayUtil;
-import org.apache.lucene.util.VectorUtil;
-import org.apache.lucene.util.hnsw.RandomVectorScorer;
-
-/**
- * Quantized vector scorer
- *
- * @lucene.experimental
- */
-public class ScalarQuantizedRandomVectorScorer
- extends RandomVectorScorer.AbstractRandomVectorScorer {
-
- public static float quantizeQuery(
- float[] query,
- byte[] quantizedQuery,
- VectorSimilarityFunction similarityFunction,
- ScalarQuantizer scalarQuantizer) {
- float[] processedQuery =
- switch (similarityFunction) {
- case EUCLIDEAN, DOT_PRODUCT, MAXIMUM_INNER_PRODUCT -> query;
- case COSINE -> {
- float[] queryCopy = ArrayUtil.copyOfSubArray(query, 0, query.length);
- VectorUtil.l2normalize(queryCopy);
- yield queryCopy;
- }
- };
- return scalarQuantizer.quantize(processedQuery, quantizedQuery, similarityFunction);
- }
-
- private final byte[] quantizedQuery;
- private final float queryOffset;
- private final RandomAccessQuantizedByteVectorValues values;
- private final ScalarQuantizedVectorSimilarity similarity;
-
- public ScalarQuantizedRandomVectorScorer(
- ScalarQuantizedVectorSimilarity similarityFunction,
- RandomAccessQuantizedByteVectorValues values,
- byte[] query,
- float queryOffset) {
- super(values);
- this.quantizedQuery = query;
- this.queryOffset = queryOffset;
- this.similarity = similarityFunction;
- this.values = values;
- }
-
- @Override
- public float score(int node) throws IOException {
- byte[] storedVectorValue = values.vectorValue(node);
- float storedVectorCorrection = values.getScoreCorrectionConstant();
- return similarity.score(
- quantizedQuery, this.queryOffset, storedVectorValue, storedVectorCorrection);
- }
-}
diff --git a/lucene/core/src/java/org/apache/lucene/util/quantization/ScalarQuantizedRandomVectorScorerSupplier.java b/lucene/core/src/java/org/apache/lucene/util/quantization/ScalarQuantizedRandomVectorScorerSupplier.java
deleted file mode 100644
index baf89df326d..00000000000
--- a/lucene/core/src/java/org/apache/lucene/util/quantization/ScalarQuantizedRandomVectorScorerSupplier.java
+++ /dev/null
@@ -1,68 +0,0 @@
-/*
- * Licensed to the Apache Software Foundation (ASF) under one or more
- * contributor license agreements. See the NOTICE file distributed with
- * this work for additional information regarding copyright ownership.
- * The ASF licenses this file to You under the Apache License, Version 2.0
- * (the "License"); you may not use this file except in compliance with
- * the License. You may obtain a copy of the License at
- *
- * http://www.apache.org/licenses/LICENSE-2.0
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-package org.apache.lucene.util.quantization;
-
-import java.io.IOException;
-import org.apache.lucene.index.VectorSimilarityFunction;
-import org.apache.lucene.util.hnsw.RandomVectorScorer;
-import org.apache.lucene.util.hnsw.RandomVectorScorerSupplier;
-
-/**
- * Quantized vector scorer supplier
- *
- * @lucene.experimental
- */
-public class ScalarQuantizedRandomVectorScorerSupplier implements RandomVectorScorerSupplier {
-
- private final RandomAccessQuantizedByteVectorValues values;
- private final ScalarQuantizedVectorSimilarity similarity;
- private final VectorSimilarityFunction vectorSimilarityFunction;
-
- public ScalarQuantizedRandomVectorScorerSupplier(
- VectorSimilarityFunction similarityFunction,
- ScalarQuantizer scalarQuantizer,
- RandomAccessQuantizedByteVectorValues values) {
- this.similarity =
- ScalarQuantizedVectorSimilarity.fromVectorSimilarity(
- similarityFunction, scalarQuantizer.getConstantMultiplier(), scalarQuantizer.getBits());
- this.values = values;
- this.vectorSimilarityFunction = similarityFunction;
- }
-
- private ScalarQuantizedRandomVectorScorerSupplier(
- ScalarQuantizedVectorSimilarity similarity,
- VectorSimilarityFunction vectorSimilarityFunction,
- RandomAccessQuantizedByteVectorValues values) {
- this.similarity = similarity;
- this.values = values;
- this.vectorSimilarityFunction = vectorSimilarityFunction;
- }
-
- @Override
- public RandomVectorScorer scorer(int ord) throws IOException {
- final RandomAccessQuantizedByteVectorValues vectorsCopy = values.copy();
- final byte[] queryVector = values.vectorValue(ord);
- final float queryOffset = values.getScoreCorrectionConstant();
- return new ScalarQuantizedRandomVectorScorer(similarity, vectorsCopy, queryVector, queryOffset);
- }
-
- @Override
- public RandomVectorScorerSupplier copy() throws IOException {
- return new ScalarQuantizedRandomVectorScorerSupplier(
- similarity, vectorSimilarityFunction, values.copy());
- }
-}
diff --git a/lucene/core/src/java21/org/apache/lucene/internal/vectorization/PanamaVectorUtilSupport.java b/lucene/core/src/java21/org/apache/lucene/internal/vectorization/PanamaVectorUtilSupport.java
index 96dafcf2c1a..9e447612215 100644
--- a/lucene/core/src/java21/org/apache/lucene/internal/vectorization/PanamaVectorUtilSupport.java
+++ b/lucene/core/src/java21/org/apache/lucene/internal/vectorization/PanamaVectorUtilSupport.java
@@ -19,7 +19,9 @@ package org.apache.lucene.internal.vectorization;
import static jdk.incubator.vector.VectorOperators.ADD;
import static jdk.incubator.vector.VectorOperators.B2I;
import static jdk.incubator.vector.VectorOperators.B2S;
+import static jdk.incubator.vector.VectorOperators.LSHR;
import static jdk.incubator.vector.VectorOperators.S2I;
+import static jdk.incubator.vector.VectorOperators.ZERO_EXTEND_B2S;
import jdk.incubator.vector.ByteVector;
import jdk.incubator.vector.FloatVector;
@@ -390,22 +392,151 @@ final class PanamaVectorUtilSupport implements VectorUtilSupport {
}
@Override
- public int int4DotProduct(byte[] a, byte[] b) {
+ public int int4DotProduct(byte[] a, boolean apacked, byte[] b, boolean bpacked) {
+ assert (apacked && bpacked) == false;
int i = 0;
int res = 0;
- if (VECTOR_BITSIZE >= 512 || VECTOR_BITSIZE == 256) {
- return dotProduct(a, b);
- } else if (a.length >= 32 && HAS_FAST_INTEGER_VECTORS) {
- i += ByteVector.SPECIES_128.loopBound(a.length);
- res += int4DotProductBody128(a, b, i);
- }
- // scalar tail
- for (; i < a.length; i++) {
- res += b[i] * a[i];
+ if (apacked || bpacked) {
+ byte[] packed = apacked ? a : b;
+ byte[] unpacked = apacked ? b : a;
+ if (packed.length >= 32) {
+ if (VECTOR_BITSIZE >= 512) {
+ i += ByteVector.SPECIES_256.loopBound(packed.length);
+ res += dotProductBody512Int4Packed(unpacked, packed, i);
+ } else if (VECTOR_BITSIZE == 256) {
+ i += ByteVector.SPECIES_128.loopBound(packed.length);
+ res += dotProductBody256Int4Packed(unpacked, packed, i);
+ } else if (HAS_FAST_INTEGER_VECTORS) {
+ i += ByteVector.SPECIES_64.loopBound(packed.length);
+ res += dotProductBody128Int4Packed(unpacked, packed, i);
+ }
+ }
+ // scalar tail
+ for (; i < packed.length; i++) {
+ byte packedByte = packed[i];
+ byte unpacked1 = unpacked[i];
+ byte unpacked2 = unpacked[i + packed.length];
+ res += (packedByte & 0x0F) * unpacked2;
+ res += ((packedByte & 0xFF) >> 4) * unpacked1;
+ }
+ } else {
+ if (VECTOR_BITSIZE >= 512 || VECTOR_BITSIZE == 256) {
+ return dotProduct(a, b);
+ } else if (a.length >= 32 && HAS_FAST_INTEGER_VECTORS) {
+ i += ByteVector.SPECIES_128.loopBound(a.length);
+ res += int4DotProductBody128(a, b, i);
+ }
+ // scalar tail
+ for (; i < a.length; i++) {
+ res += b[i] * a[i];
+ }
}
+
return res;
}
+ private int dotProductBody512Int4Packed(byte[] unpacked, byte[] packed, int limit) {
+ int sum = 0;
+ // iterate in chunks of 1024 items to ensure we don't overflow the short accumulator
+ for (int i = 0; i < limit; i += 4096) {
+ ShortVector acc0 = ShortVector.zero(ShortVector.SPECIES_512);
+ ShortVector acc1 = ShortVector.zero(ShortVector.SPECIES_512);
+ int innerLimit = Math.min(limit - i, 4096);
+ for (int j = 0; j < innerLimit; j += ByteVector.SPECIES_256.length()) {
+ // packed
+ var vb8 = ByteVector.fromArray(ByteVector.SPECIES_256, packed, i + j);
+ // unpacked
+ var va8 = ByteVector.fromArray(ByteVector.SPECIES_256, unpacked, i + j + packed.length);
+
+ // upper
+ ByteVector prod8 = vb8.and((byte) 0x0F).mul(va8);
+ Vector<Short> prod16 = prod8.convertShape(ZERO_EXTEND_B2S, ShortVector.SPECIES_512, 0);
+ acc0 = acc0.add(prod16);
+
+ // lower
+ ByteVector vc8 = ByteVector.fromArray(ByteVector.SPECIES_256, unpacked, i + j);
+ ByteVector prod8a = vb8.lanewise(LSHR, 4).mul(vc8);
+ Vector<Short> prod16a = prod8a.convertShape(ZERO_EXTEND_B2S, ShortVector.SPECIES_512, 0);
+ acc1 = acc1.add(prod16a);
+ }
+ IntVector intAcc0 = acc0.convertShape(S2I, IntVector.SPECIES_512, 0).reinterpretAsInts();
+ IntVector intAcc1 = acc0.convertShape(S2I, IntVector.SPECIES_512, 1).reinterpretAsInts();
+ IntVector intAcc2 = acc1.convertShape(S2I, IntVector.SPECIES_512, 0).reinterpretAsInts();
+ IntVector intAcc3 = acc1.convertShape(S2I, IntVector.SPECIES_512, 1).reinterpretAsInts();
+ sum += intAcc0.add(intAcc1).add(intAcc2).add(intAcc3).reduceLanes(ADD);
+ }
+ return sum;
+ }
+
+ private int dotProductBody256Int4Packed(byte[] unpacked, byte[] packed, int limit) {
+ int sum = 0;
+ // iterate in chunks of 1024 items to ensure we don't overflow the short accumulator
+ for (int i = 0; i < limit; i += 2048) {
+ ShortVector acc0 = ShortVector.zero(ShortVector.SPECIES_256);
+ ShortVector acc1 = ShortVector.zero(ShortVector.SPECIES_256);
+ int innerLimit = Math.min(limit - i, 2048);
+ for (int j = 0; j < innerLimit; j += ByteVector.SPECIES_128.length()) {
+ // packed
+ var vb8 = ByteVector.fromArray(ByteVector.SPECIES_128, packed, i + j);
+ // unpacked
+ var va8 = ByteVector.fromArray(ByteVector.SPECIES_128, unpacked, i + j + packed.length);
+
+ // upper
+ ByteVector prod8 = vb8.and((byte) 0x0F).mul(va8);
+ Vector<Short> prod16 = prod8.convertShape(ZERO_EXTEND_B2S, ShortVector.SPECIES_256, 0);
+ acc0 = acc0.add(prod16);
+
+ // lower
+ ByteVector vc8 = ByteVector.fromArray(ByteVector.SPECIES_128, unpacked, i + j);
+ ByteVector prod8a = vb8.lanewise(LSHR, 4).mul(vc8);
+ Vector<Short> prod16a = prod8a.convertShape(ZERO_EXTEND_B2S, ShortVector.SPECIES_256, 0);
+ acc1 = acc1.add(prod16a);
+ }
+ IntVector intAcc0 = acc0.convertShape(S2I, IntVector.SPECIES_256, 0).reinterpretAsInts();
+ IntVector intAcc1 = acc0.convertShape(S2I, IntVector.SPECIES_256, 1).reinterpretAsInts();
+ IntVector intAcc2 = acc1.convertShape(S2I, IntVector.SPECIES_256, 0).reinterpretAsInts();
+ IntVector intAcc3 = acc1.convertShape(S2I, IntVector.SPECIES_256, 1).reinterpretAsInts();
+ sum += intAcc0.add(intAcc1).add(intAcc2).add(intAcc3).reduceLanes(ADD);
+ }
+ return sum;
+ }
+
+ /** vectorized dot product body (128 bit vectors) */
+ private int dotProductBody128Int4Packed(byte[] unpacked, byte[] packed, int limit) {
+ int sum = 0;
+ // iterate in chunks of 1024 items to ensure we don't overflow the short accumulator
+ for (int i = 0; i < limit; i += 1024) {
+ ShortVector acc0 = ShortVector.zero(ShortVector.SPECIES_128);
+ ShortVector acc1 = ShortVector.zero(ShortVector.SPECIES_128);
+ int innerLimit = Math.min(limit - i, 1024);
+ for (int j = 0; j < innerLimit; j += ByteVector.SPECIES_64.length()) {
+ // packed
+ ByteVector vb8 = ByteVector.fromArray(ByteVector.SPECIES_64, packed, i + j);
+ // unpacked
+ ByteVector va8 =
+ ByteVector.fromArray(ByteVector.SPECIES_64, unpacked, i + j + packed.length);
+
+ // upper
+ ByteVector prod8 = vb8.and((byte) 0x0F).mul(va8);
+ ShortVector prod16 =
+ prod8.convertShape(B2S, ShortVector.SPECIES_128, 0).reinterpretAsShorts();
+ acc0 = acc0.add(prod16.and((short) 0xFF));
+
+ // lower
+ va8 = ByteVector.fromArray(ByteVector.SPECIES_64, unpacked, i + j);
+ prod8 = vb8.lanewise(LSHR, 4).mul(va8);
+ prod16 = prod8.convertShape(B2S, ShortVector.SPECIES_128, 0).reinterpretAsShorts();
+ acc1 = acc1.add(prod16.and((short) 0xFF));
+ }
+ IntVector intAcc0 = acc0.convertShape(S2I, IntVector.SPECIES_128, 0).reinterpretAsInts();
+ IntVector intAcc1 = acc0.convertShape(S2I, IntVector.SPECIES_128, 1).reinterpretAsInts();
+ IntVector intAcc2 = acc1.convertShape(S2I, IntVector.SPECIES_128, 0).reinterpretAsInts();
+ IntVector intAcc3 = acc1.convertShape(S2I, IntVector.SPECIES_128, 1).reinterpretAsInts();
+ sum += intAcc0.add(intAcc1).add(intAcc2).add(intAcc3).reduceLanes(ADD);
+ }
+ return sum;
+ }
+
private int int4DotProductBody128(byte[] a, byte[] b, int limit) {
int sum = 0;
// iterate in chunks of 1024 items to ensure we don't overflow the short accumulator
diff --git a/lucene/core/src/test/org/apache/lucene/codecs/lucene99/TestLucene99ScalarQuantizedVectorScorer.java b/lucene/core/src/test/org/apache/lucene/codecs/lucene99/TestLucene99ScalarQuantizedVectorScorer.java
new file mode 100644
index 00000000000..f926d0db7cf
--- /dev/null
+++ b/lucene/core/src/test/org/apache/lucene/codecs/lucene99/TestLucene99ScalarQuantizedVectorScorer.java
@@ -0,0 +1,155 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.lucene.codecs.lucene99;
+
+import java.io.IOException;
+import org.apache.lucene.codecs.Codec;
+import org.apache.lucene.codecs.KnnVectorsFormat;
+import org.apache.lucene.codecs.KnnVectorsReader;
+import org.apache.lucene.codecs.hnsw.DefaultFlatVectorScorer;
+import org.apache.lucene.codecs.perfield.PerFieldKnnVectorsFormat;
+import org.apache.lucene.document.Document;
+import org.apache.lucene.document.KnnFloatVectorField;
+import org.apache.lucene.index.CodecReader;
+import org.apache.lucene.index.DirectoryReader;
+import org.apache.lucene.index.IndexWriter;
+import org.apache.lucene.index.IndexWriterConfig;
+import org.apache.lucene.index.LeafReader;
+import org.apache.lucene.index.VectorSimilarityFunction;
+import org.apache.lucene.store.Directory;
+import org.apache.lucene.tests.util.LuceneTestCase;
+import org.apache.lucene.util.VectorUtil;
+import org.apache.lucene.util.hnsw.RandomVectorScorer;
+
+public class TestLucene99ScalarQuantizedVectorScorer extends LuceneTestCase {
+
+ private static Codec getCodec(int bits, boolean compress) {
+ return new Lucene99Codec() {
+ @Override
+ public KnnVectorsFormat getKnnVectorsFormatForField(String field) {
+ return new Lucene99HnswScalarQuantizedVectorsFormat(
+ Lucene99HnswVectorsFormat.DEFAULT_MAX_CONN,
+ Lucene99HnswVectorsFormat.DEFAULT_BEAM_WIDTH,
+ 1,
+ bits,
+ compress,
+ null,
+ null);
+ }
+ };
+ }
+
+ public void testScoringCompressedInt4() throws Exception {
+ vectorScoringTest(4, true);
+ }
+
+ public void testScoringUncompressedInt4() throws Exception {
+ vectorScoringTest(4, false);
+ }
+
+ public void testScoringInt7() throws Exception {
+ vectorScoringTest(7, random().nextBoolean());
+ }
+
+ private void vectorScoringTest(int bits, boolean compress) throws IOException {
+ float[][] storedVectors = new float[10][];
+ int numVectors = 10;
+ int vectorDimensions = random().nextInt(10) + 4;
+ if (bits == 4 && vectorDimensions % 2 == 1) {
+ vectorDimensions++;
+ }
+ for (int i = 0; i < numVectors; i++) {
+ float[] vector = new float[vectorDimensions];
+ for (int j = 0; j < vectorDimensions; j++) {
+ vector[j] = i + j;
+ }
+ VectorUtil.l2normalize(vector);
+ storedVectors[i] = vector;
+ }
+
+ // create lucene directory with codec
+ for (VectorSimilarityFunction similarityFunction :
+ new VectorSimilarityFunction[] {
+ VectorSimilarityFunction.DOT_PRODUCT,
+ VectorSimilarityFunction.MAXIMUM_INNER_PRODUCT,
+ VectorSimilarityFunction.EUCLIDEAN
+ }) {
+ try (Directory dir = newDirectory()) {
+ indexVectors(dir, storedVectors, similarityFunction, bits, compress);
+ try (DirectoryReader reader = DirectoryReader.open(dir)) {
+ LeafReader leafReader = reader.leaves().get(0).reader();
+ float[] vector = new float[vectorDimensions];
+ for (int i = 0; i < vectorDimensions; i++) {
+ vector[i] = i + 1;
+ }
+ VectorUtil.l2normalize(vector);
+ RandomVectorScorer randomScorer =
+ getRandomVectorScorer(similarityFunction, leafReader, vector);
+ float[] rawScores = new float[10];
+ for (int i = 0; i < 10; i++) {
+ rawScores[i] = similarityFunction.compare(vector, storedVectors[i]);
+ }
+ for (int i = 0; i < 10; i++) {
+ assertEquals(similarityFunction.toString(), rawScores[i], randomScorer.score(i), 0.05f);
+ }
+ }
+ }
+ }
+ }
+
+ private RandomVectorScorer getRandomVectorScorer(
+ VectorSimilarityFunction function, LeafReader leafReader, float[] vector) throws IOException {
+ if (leafReader instanceof CodecReader codecReader) {
+ KnnVectorsReader format = codecReader.getVectorReader();
+ if (format instanceof PerFieldKnnVectorsFormat.FieldsReader perFieldFormat) {
+ format = perFieldFormat.getFieldReader("field");
+ }
+ if (format instanceof Lucene99HnswVectorsReader hnswFormat) {
+ OffHeapQuantizedByteVectorValues quantizedByteVectorReader =
+ (OffHeapQuantizedByteVectorValues) hnswFormat.getQuantizedVectorValues("field");
+ return new Lucene99ScalarQuantizedVectorScorer(new DefaultFlatVectorScorer())
+ .getRandomVectorScorer(function, quantizedByteVectorReader, vector);
+ }
+ }
+ throw new IllegalArgumentException("Unsupported reader");
+ }
+
+ private static void indexVectors(
+ Directory dir,
+ float[][] vectors,
+ VectorSimilarityFunction function,
+ int bits,
+ boolean compress)
+ throws IOException {
+ try (IndexWriter writer =
+ new IndexWriter(dir, new IndexWriterConfig().setCodec(getCodec(bits, compress)))) {
+ for (int i = 0; i < vectors.length; i++) {
+ Document doc = new Document();
+ if (random().nextBoolean()) {
+ // index a document without a vector
+ writer.addDocument(doc);
+ }
+ writer.addDocument(doc);
+ doc.add(new KnnFloatVectorField("field", vectors[i], function));
+ writer.addDocument(doc);
+ }
+ writer.commit();
+ writer.forceMerge(1);
+ }
+ }
+}
diff --git a/lucene/core/src/test/org/apache/lucene/internal/vectorization/TestVectorUtilSupport.java b/lucene/core/src/test/org/apache/lucene/internal/vectorization/TestVectorUtilSupport.java
index 9fe5ddd0e2b..7064955cb5f 100644
--- a/lucene/core/src/test/org/apache/lucene/internal/vectorization/TestVectorUtilSupport.java
+++ b/lucene/core/src/test/org/apache/lucene/internal/vectorization/TestVectorUtilSupport.java
@@ -27,7 +27,8 @@ public class TestVectorUtilSupport extends BaseVectorizationTestCase {
private static final double DELTA = 1e-3;
private static final int[] VECTOR_SIZES = {
- 1, 4, 6, 8, 13, 16, 25, 32, 64, 100, 128, 207, 256, 300, 512, 702, 1024
+ 1, 4, 6, 8, 13, 16, 25, 32, 64, 100, 128, 207, 256, 300, 512, 702, 1024, 1536, 2046, 2048, 4096,
+ 4098
};
private final int size;
@@ -92,6 +93,55 @@ public class TestVectorUtilSupport extends BaseVectorizationTestCase {
assertFloatReturningProviders(p -> p.cosine(a, b));
}
+ public void testInt4DotProduct() {
+ assumeTrue("even sizes only", size % 2 == 0);
+ var a = new byte[size];
+ var b = new byte[size];
+ for (int i = 0; i < size; ++i) {
+ a[i] = (byte) random().nextInt(16);
+ b[i] = (byte) random().nextInt(16);
+ }
+
+ assertIntReturningProviders(p -> p.int4DotProduct(a, false, pack(b), true));
+ assertIntReturningProviders(p -> p.int4DotProduct(pack(a), true, b, false));
+ assertEquals(
+ LUCENE_PROVIDER.getVectorUtilSupport().dotProduct(a, b),
+ PANAMA_PROVIDER.getVectorUtilSupport().int4DotProduct(a, false, pack(b), true));
+ }
+
+ public void testInt4DotProductBoundaries() {
+ assumeTrue("even sizes only", size % 2 == 0);
+ byte MAX_VALUE = 15;
+ var a = new byte[size];
+ var b = new byte[size];
+
+ Arrays.fill(a, MAX_VALUE);
+ Arrays.fill(b, MAX_VALUE);
+ assertIntReturningProviders(p -> p.int4DotProduct(a, false, pack(b), true));
+ assertIntReturningProviders(p -> p.int4DotProduct(pack(a), true, b, false));
+ assertEquals(
+ LUCENE_PROVIDER.getVectorUtilSupport().dotProduct(a, b),
+ PANAMA_PROVIDER.getVectorUtilSupport().int4DotProduct(a, false, pack(b), true));
+
+ byte MIN_VALUE = 0;
+ Arrays.fill(a, MIN_VALUE);
+ Arrays.fill(b, MIN_VALUE);
+ assertIntReturningProviders(p -> p.int4DotProduct(a, false, pack(b), true));
+ assertIntReturningProviders(p -> p.int4DotProduct(pack(a), true, b, false));
+ assertEquals(
+ LUCENE_PROVIDER.getVectorUtilSupport().dotProduct(a, b),
+ PANAMA_PROVIDER.getVectorUtilSupport().int4DotProduct(a, false, pack(b), true));
+ }
+
+ static byte[] pack(byte[] unpacked) {
+ int len = (unpacked.length + 1) / 2;
+ var packed = new byte[len];
+ for (int i = 0; i < len; i++) {
+ packed[i] = (byte) (unpacked[i] << 4 | unpacked[packed.length + i]);
+ }
+ return packed;
+ }
+
private void assertFloatReturningProviders(ToDoubleFunction<VectorUtilSupport> func) {
assertEquals(
func.applyAsDouble(LUCENE_PROVIDER.getVectorUtilSupport()),