io.druid.benchmark.indexing.IndexPersistBenchmark.java Source code

Introduction

Here is the source code for io.druid.benchmark.indexing.IndexPersistBenchmark.java

Source

/*
 * Licensed to Metamarkets Group Inc. (Metamarkets) under one
 * or more contributor license agreements. See the NOTICE file
 * distributed with this work for additional information
 * regarding copyright ownership. Metamarkets 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 io.druid.benchmark.indexing;

import com.fasterxml.jackson.databind.ObjectMapper;
import com.google.common.io.Files;
import io.druid.benchmark.datagen.BenchmarkDataGenerator;
import io.druid.benchmark.datagen.BenchmarkSchemaInfo;
import io.druid.benchmark.datagen.BenchmarkSchemas;
import io.druid.data.input.InputRow;
import io.druid.hll.HyperLogLogHash;
import io.druid.jackson.DefaultObjectMapper;
import io.druid.java.util.common.logger.Logger;
import io.druid.query.aggregation.hyperloglog.HyperUniquesSerde;
import io.druid.segment.IndexIO;
import io.druid.segment.IndexMergerV9;
import io.druid.segment.IndexSpec;
import io.druid.segment.column.ColumnConfig;
import io.druid.segment.incremental.IncrementalIndex;
import io.druid.segment.incremental.IncrementalIndexSchema;
import io.druid.segment.serde.ComplexMetrics;
import org.apache.commons.io.FileUtils;
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.Warmup;
import org.openjdk.jmh.infra.Blackhole;

import java.io.File;
import java.io.IOException;
import java.util.ArrayList;
import java.util.concurrent.TimeUnit;

@State(Scope.Benchmark)
@Fork(value = 1)
@Warmup(iterations = 10)
@Measurement(iterations = 25)
public class IndexPersistBenchmark {
    @Param({ "75000" })
    private int rowsPerSegment;

    @Param({ "basic" })
    private String schema;

    @Param({ "true", "false" })
    private boolean rollup;

    private static final Logger log = new Logger(IndexPersistBenchmark.class);
    private static final int RNG_SEED = 9999;

    private IncrementalIndex incIndex;
    private ArrayList<InputRow> rows;
    private BenchmarkSchemaInfo schemaInfo;

    private static final IndexMergerV9 INDEX_MERGER_V9;
    private static final IndexIO INDEX_IO;
    public static final ObjectMapper JSON_MAPPER;

    static {
        JSON_MAPPER = new DefaultObjectMapper();
        INDEX_IO = new IndexIO(JSON_MAPPER, new ColumnConfig() {
            @Override
            public int columnCacheSizeBytes() {
                return 0;
            }
        });
        INDEX_MERGER_V9 = new IndexMergerV9(JSON_MAPPER, INDEX_IO);
    }

    @Setup
    public void setup() throws IOException {
        log.info("SETUP CALLED AT " + System.currentTimeMillis());

        if (ComplexMetrics.getSerdeForType("hyperUnique") == null) {
            ComplexMetrics.registerSerde("hyperUnique", new HyperUniquesSerde(HyperLogLogHash.getDefault()));
        }

        rows = new ArrayList<InputRow>();
        schemaInfo = BenchmarkSchemas.SCHEMA_MAP.get(schema);

        BenchmarkDataGenerator gen = new BenchmarkDataGenerator(schemaInfo.getColumnSchemas(), RNG_SEED,
                schemaInfo.getDataInterval(), rowsPerSegment);

        for (int i = 0; i < rowsPerSegment; i++) {
            InputRow row = gen.nextRow();
            if (i % 10000 == 0) {
                log.info(i + " rows generated.");
            }
            rows.add(row);
        }

    }

    @Setup(Level.Iteration)
    public void setup2() throws IOException {
        incIndex = makeIncIndex();
        for (int i = 0; i < rowsPerSegment; i++) {
            InputRow row = rows.get(i);
            incIndex.add(row);
        }
    }

    @TearDown(Level.Iteration)
    public void teardown() throws IOException {
        incIndex.close();
        incIndex = null;
    }

    private IncrementalIndex makeIncIndex() {
        return new IncrementalIndex.Builder()
                .setIndexSchema(new IncrementalIndexSchema.Builder().withMetrics(schemaInfo.getAggsArray())
                        .withRollup(rollup).build())
                .setReportParseExceptions(false).setMaxRowCount(rowsPerSegment).buildOnheap();
    }

    @Benchmark
    @BenchmarkMode(Mode.AverageTime)
    @OutputTimeUnit(TimeUnit.MICROSECONDS)
    public void persistV9(Blackhole blackhole) throws Exception {
        File tmpDir = Files.createTempDir();
        log.info("Using temp dir: " + tmpDir.getAbsolutePath());
        try {
            File indexFile = INDEX_MERGER_V9.persist(incIndex, tmpDir, new IndexSpec());

            blackhole.consume(indexFile);

        } finally {
            FileUtils.deleteDirectory(tmpDir);
        }
    }
}