Merge branch 'main' into eis-text-embedding-task-type

Author: brendan-jugan-elastic

benchmarks/src/main/java/org/elasticsearch/benchmark/vector/Int4ScorerBenchmark.java

@@ -0,0 +1,123 @@
+/*
+ * Copyright Elasticsearch B.V. and/or licensed to Elasticsearch B.V. under one
+ * or more contributor license agreements. Licensed under the "Elastic License
+ * 2.0", the "GNU Affero General Public License v3.0 only", and the "Server Side
+ * Public License v 1"; you may not use this file except in compliance with, at
+ * your election, the "Elastic License 2.0", the "GNU Affero General Public
+ * License v3.0 only", or the "Server Side Public License, v 1".
+ */
+package org.elasticsearch.benchmark.vector;
+
+import org.apache.lucene.store.Directory;
+import org.apache.lucene.store.IOContext;
+import org.apache.lucene.store.IndexInput;
+import org.apache.lucene.store.IndexOutput;
+import org.apache.lucene.store.MMapDirectory;
+import org.apache.lucene.util.VectorUtil;
+import org.elasticsearch.common.logging.LogConfigurator;
+import org.elasticsearch.core.IOUtils;
+import org.elasticsearch.simdvec.ES91Int4VectorsScorer;
+import org.elasticsearch.simdvec.internal.vectorization.ESVectorizationProvider;
+import org.openjdk.jmh.annotations.Benchmark;
+import org.openjdk.jmh.annotations.BenchmarkMode;
+import org.openjdk.jmh.annotations.Fork;
+import org.openjdk.jmh.annotations.Measurement;
+import org.openjdk.jmh.annotations.Mode;
+import org.openjdk.jmh.annotations.OutputTimeUnit;
+import org.openjdk.jmh.annotations.Param;
+import org.openjdk.jmh.annotations.Scope;
+import org.openjdk.jmh.annotations.Setup;
+import org.openjdk.jmh.annotations.State;
+import org.openjdk.jmh.annotations.TearDown;
+import org.openjdk.jmh.annotations.Warmup;
+import org.openjdk.jmh.infra.Blackhole;
+
+import java.io.IOException;
+import java.nio.file.Files;
+import java.util.concurrent.ThreadLocalRandom;
+import java.util.concurrent.TimeUnit;
+
+@BenchmarkMode(Mode.Throughput)
+@OutputTimeUnit(TimeUnit.MILLISECONDS)
+@State(Scope.Benchmark)
+// first iteration is complete garbage, so make sure we really warmup
+@Warmup(iterations = 4, time = 1)
+// real iterations. not useful to spend tons of time here, better to fork more
+@Measurement(iterations = 5, time = 1)
+// engage some noise reduction
+@Fork(value = 1)
+public class Int4ScorerBenchmark {
+
+    static {
+        LogConfigurator.configureESLogging(); // native access requires logging to be initialized
+    }
+
+    @Param({ "384", "702", "1024" })
+    int dims;
+
+    int numVectors = 200;
+    int numQueries = 10;
+
+    byte[] scratch;
+    byte[][] binaryVectors;
+    byte[][] binaryQueries;
+
+    ES91Int4VectorsScorer scorer;
+    Directory dir;
+    IndexInput in;
+
+    @Setup
+    public void setup() throws IOException {
+        binaryVectors = new byte[numVectors][dims];
+        dir = new MMapDirectory(Files.createTempDirectory("vectorData"));
+        try (IndexOutput out = dir.createOutput("vectors", IOContext.DEFAULT)) {
+            for (byte[] binaryVector : binaryVectors) {
+                for (int i = 0; i < dims; i++) {
+                    // 4-bit quantization
+                    binaryVector[i] = (byte) ThreadLocalRandom.current().nextInt(16);
+                }
+                out.writeBytes(binaryVector, 0, binaryVector.length);
+            }
+        }
+
+        in = dir.openInput("vectors", IOContext.DEFAULT);
+        binaryQueries = new byte[numVectors][dims];
+        for (byte[] binaryVector : binaryVectors) {
+            for (int i = 0; i < dims; i++) {
+                // 4-bit quantization
+                binaryVector[i] = (byte) ThreadLocalRandom.current().nextInt(16);
+            }
+        }
+
+        scratch = new byte[dims];
+        scorer = ESVectorizationProvider.getInstance().newES91Int4VectorsScorer(in, dims);
+    }
+
+    @TearDown
+    public void teardown() throws IOException {
+        IOUtils.close(dir, in);
+    }
+
+    @Benchmark
+    @Fork(jvmArgsPrepend = { "--add-modules=jdk.incubator.vector" })
+    public void scoreFromArray(Blackhole bh) throws IOException {
+        for (int j = 0; j < numQueries; j++) {
+            in.seek(0);
+            for (int i = 0; i < numVectors; i++) {
+                in.readBytes(scratch, 0, dims);
+                bh.consume(VectorUtil.int4DotProduct(binaryQueries[j], scratch));
+            }
+        }
+    }
+
+    @Benchmark
+    @Fork(jvmArgsPrepend = { "--add-modules=jdk.incubator.vector" })
+    public void scoreFromMemorySegmentOnlyVector(Blackhole bh) throws IOException {
+        for (int j = 0; j < numQueries; j++) {
+            in.seek(0);
+            for (int i = 0; i < numVectors; i++) {
+                bh.consume(scorer.int4DotProduct(binaryQueries[j]));
+            }
+        }
+    }
+}

docs/changelog/125143.yaml

@@ -0,0 +1,6 @@
+pr: 129325
+summary: Check for model deployment in inference endpoints before stopping
+area: Machine Learning
+type: bug
+issues:
+ - 128549

docs/changelog/129325.yaml

@@ -0,0 +1,5 @@
+pr: 129659
+summary: Simplified RRF Retriever
+area: Search
+type: enhancement
+issues: []

docs/changelog/129659.yaml

@@ -10,12 +10,3 @@
   * [preview] [`ST_XMIN`](../../functions-operators/spatial-functions.md#esql-st_xmin)
   * [preview] [`ST_YMAX`](../../functions-operators/spatial-functions.md#esql-st_ymax)
   * [preview] [`ST_YMIN`](../../functions-operators/spatial-functions.md#esql-st_ymin)
-* [preview] [`ST_GEOTILE`](../../functions-operators/spatial-functions.md#esql-st_geotile)
-  * [preview] [`ST_GEOTILE_TO_STRING`](../../functions-operators/spatial-functions.md#esql-st_geotile_to_string)
-  * [preview] [`ST_GEOTILE_TO_LONG`](../../functions-operators/spatial-functions.md#esql-st_geotile_to_long)
-* [preview] [`ST_GEOHEX`](../../functions-operators/spatial-functions.md#esql-st_geohex)
-  * [preview] [`ST_GEOHEX_TO_STRING`](../../functions-operators/spatial-functions.md#esql-st_geohex_to_string)
-  * [preview] [`ST_GEOHEX_TO_LONG`](../../functions-operators/spatial-functions.md#esql-st_geohex_to_long)
-* [preview] [`ST_GEOHASH`](../../functions-operators/spatial-functions.md#esql-st_geohash)
-  * [preview] [`ST_GEOHASH_TO_STRING`](../../functions-operators/spatial-functions.md#esql-st_geohash_to_string)
-  * [preview] [`ST_GEOHASH_TO_LONG`](../../functions-operators/spatial-functions.md#esql-st_geohash_to_long)

docs/reference/query-languages/esql/_snippets/lists/spatial-functions.md

@@ -48,29 +48,3 @@ mapped_pages:
 :::{include} ../_snippets/functions/layout/st_ymin.md
 :::
 
-:::{include} ../_snippets/functions/layout/st_geotile.md
-:::
-
-:::{include} ../_snippets/functions/layout/st_geotile_to_string.md
-:::
-
-:::{include} ../_snippets/functions/layout/st_geotile_to_long.md
-:::
-
-:::{include} ../_snippets/functions/layout/st_geohex.md
-:::
-
-:::{include} ../_snippets/functions/layout/st_geohex_to_string.md
-:::
-
-:::{include} ../_snippets/functions/layout/st_geohex_to_long.md
-:::
-
-:::{include} ../_snippets/functions/layout/st_geohash.md
-:::
-
-:::{include} ../_snippets/functions/layout/st_geohash_to_string.md
-:::
-
-:::{include} ../_snippets/functions/layout/st_geohash_to_long.md
-:::