Merge branch 'main' into remove-empty-attribute-in-merging-output

Author: kanoshiou

.buildkite/pipelines/pull-request/build-benchmark.yml

@@ -7,16 +7,19 @@ steps:
     command: |
       .ci/scripts/run-gradle.sh :build-tools-internal:bootstrapPerformanceTests
       .ci/scripts/install-gradle-profiler.sh
-      .ci/scripts/run-gradle-profiler.sh --benchmark --scenario-file build-tools-internal/build/performanceTests/elasticsearch-build-benchmark-{{matrix.part}}.scenarios --project-dir . --output-dir profile-out
-      mkdir build
-      tar -czf build/$BUILDKITE_BUILD_NUMBER.tar.bz2 profile-out
+      .ci/scripts/run-gradle-profiler.sh --benchmark --scenario-file build-tools-internal/build/performanceTests/elasticsearch-{{matrix.part}}.scenarios --measure-config-time --project-dir . --output-dir profile-out
+      mkdir -p build
+      buildkite-agent artifact upload "profile-out/*"
+      cat << EOF | buildkite-agent annotate --style "info"
+      Benchmark Html Report <a href="artifact://profile-out/benchmark.html">Benchmark Html Report</a>
+      EOF
     matrix:
       setup:
         part:
-          - part1
-          - part2
+          - esql-tests
     env:
       BUILD_PERFORMANCE_TEST: "true"
+
     agents:
       provider: gcp
       image: family/elasticsearch-ubuntu-2004

.ci/scripts/install-gradle-profiler.sh

@@ -2,7 +2,7 @@
 
 set -e
 # profiler version we wanna install
-PROFILER_VERSION="0.16.0"
-wget https://repo.gradle.org/gradle/ext-releases-local/org/gradle/profiler/gradle-profiler/$PROFILER_VERSION/gradle-profiler-$PROFILER_VERSION.zip -O $WORKSPACE/gradle-profiler-$PROFILER_VERSION.zip
+PROFILER_VERSION="0.22.0"
+wget https://repo1.maven.org/maven2/org/gradle/profiler/gradle-profiler/$PROFILER_VERSION/gradle-profiler-$PROFILER_VERSION.zip -O $WORKSPACE/gradle-profiler-$PROFILER_VERSION.zip
 unzip $WORKSPACE/gradle-profiler-$PROFILER_VERSION.zip
 mv $WORKSPACE/gradle-profiler-$PROFILER_VERSION $WORKSPACE/gradle-profiler

benchmarks/README.md

@@ -82,19 +82,21 @@ To get realistic results, you should exercise care when running benchmarks. Here
 NOTE: Linux only. Sorry Mac and Windows.
 
 Disassembling is fun! Maybe not always useful, but always fun! Generally, you'll want to install `perf` and the JDK's `hsdis`.
-`perf` is generally available via `apg-get install perf` or `pacman -S perf`. `hsdis` you'll want to compile from source. is a little more involved. This worked
+`perf` is generally available via `apg-get install perf` or `pacman -S perf linux-tools`. `hsdis` you'll want to compile from source. is a little more involved. This worked
 on 2020-08-01:
 
 ```
 git clone [email protected]:openjdk/jdk.git
 cd jdk
-git checkout jdk-17-ga
-cd src/utils/hsdis
+git checkout jdk-24-ga
 # Get a known good binutils
 wget https://ftp.gnu.org/gnu/binutils/binutils-2.35.tar.gz
 tar xf binutils-2.35.tar.gz
-make BINUTILS=binutils-2.35 ARCH=amd64
-sudo cp build/linux-amd64/hsdis-amd64.so /usr/lib/jvm/java-17-openjdk/lib/server/
+bash configure --with-hsdis=binutils --with-binutils-src=binutils-2.35 \
+    --with-boot-jdk=~/.gradle/jdks/oracle_corporation-24-amd64-linux.2
+make build-hsdis
+cp ./build/linux-x86_64-server-release/jdk/lib/hsdis-amd64.so \
+    ~/.gradle/jdks/oracle_corporation-24-amd64-linux.2/lib/hsdis.so
 ```
 
 If you want to disassemble a single method do something like this:
@@ -105,6 +107,30 @@ gradlew -p benchmarks run --args ' MemoryStatsBenchmark -jvmArgs "-XX:+UnlockDia
 
 If you want `perf` to find the hot methods for you, then do add `-prof perfasm`.
 
+NOTE: `perfasm` will need more access:
+```
+sudo bash
+echo -1 > /proc/sys/kernel/perf_event_paranoid
+exit
+```
+
+If you get warnings like:
+```
+The perf event count is suspiciously low (0).
+```
+then check if you are bumping into [this](https://man.archlinux.org/man/perf-stat.1.en#INTEL_HYBRID_SUPPORT)
+by running:
+```
+perf stat -B dd if=/dev/zero of=/dev/null count=1000000
+```
+
+If you see lines like:
+```
+         765019980      cpu_atom/cycles/                 #    1.728 GHz                         (0.60%)
+        2258845959      cpu_core/cycles/                 #    5.103 GHz                         (99.18%)
+```
+then `perf` is just not going to work for you.
+
 ## Async Profiler
 
 Note: Linux and Mac only. Sorry Windows.

benchmarks/src/main/java/org/elasticsearch/benchmark/compute/operator/ParseIpBenchmark.java

@@ -0,0 +1,61 @@
+/*
+ * 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.compute.operator;
+
+import org.apache.lucene.document.InetAddressPoint;
+import org.apache.lucene.util.BytesRef;
+import org.elasticsearch.common.breaker.NoopCircuitBreaker;
+import org.elasticsearch.common.network.InetAddresses;
+import org.elasticsearch.compute.operator.BreakingBytesRefBuilder;
+import org.elasticsearch.xpack.esql.expression.function.scalar.convert.ParseIp;
+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.Scope;
+import org.openjdk.jmh.annotations.State;
+import org.openjdk.jmh.annotations.Warmup;
+
+import java.net.InetAddress;
+import java.util.concurrent.TimeUnit;
+
+@Warmup(iterations = 5)
+@Measurement(iterations = 7)
+@BenchmarkMode(Mode.AverageTime)
+@OutputTimeUnit(TimeUnit.NANOSECONDS)
+@State(Scope.Thread)
+@Fork(1)
+public class ParseIpBenchmark {
+    private final BytesRef ip = new BytesRef("192.168.0.1");
+    private final BreakingBytesRefBuilder scratch = ParseIp.buildScratch(new NoopCircuitBreaker("request"));
+
+    @Benchmark
+    public BytesRef leadingZerosRejected() {
+        return ParseIp.leadingZerosRejected(ip, scratch);
+    }
+
+    @Benchmark
+    public BytesRef leadingZerosAreDecimal() {
+        return ParseIp.leadingZerosAreDecimal(ip, scratch);
+    }
+
+    @Benchmark
+    public BytesRef leadingZerosAreOctal() {
+        return ParseIp.leadingZerosAreOctal(ip, scratch);
+    }
+
+    @Benchmark
+    public BytesRef original() {
+        InetAddress inetAddress = InetAddresses.forString(ip.utf8ToString());
+        return new BytesRef(InetAddressPoint.encode(inetAddress));
+    }
+}

benchmarks/src/main/java/org/elasticsearch/benchmark/index/codec/tsdb/TSDBDocValuesMergeBenchmark.java

@@ -0,0 +1,196 @@
+/*
+ * 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.index.codec.tsdb;
+
+import org.apache.lucene.analysis.standard.StandardAnalyzer;
+import org.apache.lucene.codecs.DocValuesFormat;
+import org.apache.lucene.document.Document;
+import org.apache.lucene.document.SortedDocValuesField;
+import org.apache.lucene.document.SortedNumericDocValuesField;
+import org.apache.lucene.document.SortedSetDocValuesField;
+import org.apache.lucene.index.IndexWriter;
+import org.apache.lucene.index.IndexWriterConfig;
+import org.apache.lucene.index.LogByteSizeMergePolicy;
+import org.apache.lucene.search.Sort;
+import org.apache.lucene.search.SortField;
+import org.apache.lucene.search.SortedNumericSortField;
+import org.apache.lucene.store.Directory;
+import org.apache.lucene.store.FSDirectory;
+import org.apache.lucene.util.BytesRef;
+import org.elasticsearch.cluster.metadata.DataStream;
+import org.elasticsearch.common.logging.LogConfigurator;
+import org.elasticsearch.index.codec.Elasticsearch900Lucene101Codec;
+import org.elasticsearch.index.codec.tsdb.es819.ES819TSDBDocValuesFormat;
+import org.openjdk.jmh.annotations.Benchmark;
+import org.openjdk.jmh.annotations.BenchmarkMode;
+import org.openjdk.jmh.annotations.Fork;
+import org.openjdk.jmh.annotations.Level;
+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.Threads;
+import org.openjdk.jmh.annotations.Warmup;
+import org.openjdk.jmh.profile.AsyncProfiler;
+import org.openjdk.jmh.runner.Runner;
+import org.openjdk.jmh.runner.RunnerException;
+import org.openjdk.jmh.runner.options.Options;
+import org.openjdk.jmh.runner.options.OptionsBuilder;
+
+import java.io.IOException;
+import java.nio.file.Files;
+import java.util.Random;
+import java.util.concurrent.ExecutorService;
+import java.util.concurrent.Executors;
+import java.util.concurrent.TimeUnit;
+
+@BenchmarkMode(Mode.SampleTime)
+@OutputTimeUnit(TimeUnit.MILLISECONDS)
+@State(Scope.Benchmark)
+@Fork(1)
+@Threads(1)
+@Warmup(iterations = 0)
+@Measurement(iterations = 1)
+public class TSDBDocValuesMergeBenchmark {
+
+    static {
+        // For Elasticsearch900Lucene101Codec:
+        LogConfigurator.loadLog4jPlugins();
+        LogConfigurator.configureESLogging();
+        LogConfigurator.setNodeName("test");
+    }
+
+    @Param("20431204")
+    private int nDocs;
+
+    @Param("1000")
+    private int deltaTime;
+
+    @Param("42")
+    private int seed;
+
+    private static final String TIMESTAMP_FIELD = "@timestamp";
+    private static final String HOSTNAME_FIELD = "host.name";
+    private static final long BASE_TIMESTAMP = 1704067200000L;
+
+    private IndexWriter indexWriterWithoutOptimizedMerge;
+    private IndexWriter indexWriterWithOptimizedMerge;
+    private ExecutorService executorService;
+
+    public static void main(String[] args) throws RunnerException {
+        final Options options = new OptionsBuilder().include(TSDBDocValuesMergeBenchmark.class.getSimpleName())
+            .addProfiler(AsyncProfiler.class)
+            .build();
+
+        new Runner(options).run();
+    }
+
+    @Setup(Level.Trial)
+    public void setup() throws IOException {
+        executorService = Executors.newSingleThreadExecutor();
+
+        final Directory tempDirectoryWithoutDocValuesSkipper = FSDirectory.open(Files.createTempDirectory("temp1-"));
+        final Directory tempDirectoryWithDocValuesSkipper = FSDirectory.open(Files.createTempDirectory("temp2-"));
+
+        indexWriterWithoutOptimizedMerge = createIndex(tempDirectoryWithoutDocValuesSkipper, false);
+        indexWriterWithOptimizedMerge = createIndex(tempDirectoryWithDocValuesSkipper, true);
+    }
+
+    private IndexWriter createIndex(final Directory directory, final boolean optimizedMergeEnabled) throws IOException {
+        final var iwc = createIndexWriterConfig(optimizedMergeEnabled);
+        long counter1 = 0;
+        long counter2 = 10_000_000;
+        long[] gauge1Values = new long[] { 2, 4, 6, 8, 10, 12, 14, 16 };
+        long[] gauge2Values = new long[] { -2, -4, -6, -8, -10, -12, -14, -16 };
+        int numHosts = 1000;
+        String[] tags = new String[] { "tag_1", "tag_2", "tag_3", "tag_4", "tag_5", "tag_6", "tag_7", "tag_8" };
+
+        final Random random = new Random(seed);
+        IndexWriter indexWriter = new IndexWriter(directory, iwc);
+        for (int i = 0; i < nDocs; i++) {
+            final Document doc = new Document();
+
+            final int batchIndex = i / numHosts;
+            final String hostName = "host-" + batchIndex;
+            // Slightly vary the timestamp in each document
+            final long timestamp = BASE_TIMESTAMP + ((i % numHosts) * deltaTime) + random.nextInt(0, deltaTime);
+
+            doc.add(new SortedDocValuesField(HOSTNAME_FIELD, new BytesRef(hostName)));
+            doc.add(new SortedNumericDocValuesField(TIMESTAMP_FIELD, timestamp));
+            doc.add(new SortedNumericDocValuesField("counter_1", counter1++));
+            doc.add(new SortedNumericDocValuesField("counter_2", counter2++));
+            doc.add(new SortedNumericDocValuesField("gauge_1", gauge1Values[i % gauge1Values.length]));
+            doc.add(new SortedNumericDocValuesField("gauge_2", gauge2Values[i % gauge1Values.length]));
+            int numTags = tags.length % (i + 1);
+            for (int j = 0; j < numTags; j++) {
+                doc.add(new SortedSetDocValuesField("tags", new BytesRef(tags[j])));
+            }
+
+            indexWriter.addDocument(doc);
+        }
+        indexWriter.commit();
+        return indexWriter;
+    }
+
+    @Benchmark
+    public void forceMergeWithoutOptimizedMerge() throws IOException {
+        forceMerge(indexWriterWithoutOptimizedMerge);
+    }
+
+    @Benchmark
+    public void forceMergeWithOptimizedMerge() throws IOException {
+        forceMerge(indexWriterWithOptimizedMerge);
+    }
+
+    private void forceMerge(final IndexWriter indexWriter) throws IOException {
+        indexWriter.forceMerge(1);
+    }
+
+    @TearDown(Level.Trial)
+    public void tearDown() {
+        if (executorService != null) {
+            executorService.shutdown();
+            try {
+                if (executorService.awaitTermination(30, TimeUnit.SECONDS) == false) {
+                    executorService.shutdownNow();
+                }
+            } catch (InterruptedException e) {
+                executorService.shutdownNow();
+                Thread.currentThread().interrupt();
+            }
+        }
+    }
+
+    private static IndexWriterConfig createIndexWriterConfig(boolean optimizedMergeEnabled) {
+        var config = new IndexWriterConfig(new StandardAnalyzer());
+        // NOTE: index sort config matching LogsDB's sort order
+        config.setIndexSort(
+            new Sort(
+                new SortField(HOSTNAME_FIELD, SortField.Type.STRING, false),
+                new SortedNumericSortField(TIMESTAMP_FIELD, SortField.Type.LONG, true)
+            )
+        );
+        config.setLeafSorter(DataStream.TIMESERIES_LEAF_READERS_SORTER);
+        config.setMergePolicy(new LogByteSizeMergePolicy());
+        var docValuesFormat = new ES819TSDBDocValuesFormat(4096, optimizedMergeEnabled);
+        config.setCodec(new Elasticsearch900Lucene101Codec() {
+
+            @Override
+            public DocValuesFormat getDocValuesFormatForField(String field) {
+                return docValuesFormat;
+            }
+        });
+        return config;
+    }
+}