Search in sources :

Example 6 with BenchmarkDataGenerator

use of io.druid.benchmark.datagen.BenchmarkDataGenerator in project druid by druid-io.

the class SearchBenchmark method setup.

@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()));
    }
    executorService = Execs.multiThreaded(numSegments, "SearchThreadPool");
    setupQueries();
    String[] schemaQuery = schemaAndQuery.split("\\.");
    String schemaName = schemaQuery[0];
    String queryName = schemaQuery[1];
    schemaInfo = BenchmarkSchemas.SCHEMA_MAP.get(schemaName);
    queryBuilder = SCHEMA_QUERY_MAP.get(schemaName).get(queryName);
    queryBuilder.limit(limit);
    query = queryBuilder.build();
    incIndexes = new ArrayList<>();
    for (int i = 0; i < numSegments; i++) {
        log.info("Generating rows for segment " + i);
        BenchmarkDataGenerator gen = new BenchmarkDataGenerator(schemaInfo.getColumnSchemas(), System.currentTimeMillis(), schemaInfo.getDataInterval(), rowsPerSegment);
        IncrementalIndex incIndex = makeIncIndex();
        for (int j = 0; j < rowsPerSegment; j++) {
            InputRow row = gen.nextRow();
            if (j % 10000 == 0) {
                log.info(j + " rows generated.");
            }
            incIndex.add(row);
        }
        incIndexes.add(incIndex);
    }
    tmpDir = Files.createTempDir();
    log.info("Using temp dir: " + tmpDir.getAbsolutePath());
    qIndexes = new ArrayList<>();
    for (int i = 0; i < numSegments; i++) {
        File indexFile = INDEX_MERGER_V9.persist(incIndexes.get(i), tmpDir, new IndexSpec());
        QueryableIndex qIndex = INDEX_IO.loadIndex(indexFile);
        qIndexes.add(qIndex);
    }
    final SearchQueryConfig config = new SearchQueryConfig().withOverrides(query);
    factory = new SearchQueryRunnerFactory(new SearchStrategySelector(Suppliers.ofInstance(config)), new SearchQueryQueryToolChest(config, QueryBenchmarkUtil.NoopIntervalChunkingQueryRunnerDecorator()), QueryBenchmarkUtil.NOOP_QUERYWATCHER);
}
Also used : IndexSpec(io.druid.segment.IndexSpec) SearchQueryConfig(io.druid.query.search.search.SearchQueryConfig) IncrementalIndex(io.druid.segment.incremental.IncrementalIndex) OnheapIncrementalIndex(io.druid.segment.incremental.OnheapIncrementalIndex) BenchmarkDataGenerator(io.druid.benchmark.datagen.BenchmarkDataGenerator) HyperUniquesSerde(io.druid.query.aggregation.hyperloglog.HyperUniquesSerde) SearchQueryRunnerFactory(io.druid.query.search.SearchQueryRunnerFactory) SearchStrategySelector(io.druid.query.search.SearchStrategySelector) SearchQueryQueryToolChest(io.druid.query.search.SearchQueryQueryToolChest) QueryableIndex(io.druid.segment.QueryableIndex) InputRow(io.druid.data.input.InputRow) File(java.io.File) Setup(org.openjdk.jmh.annotations.Setup)

Example 7 with BenchmarkDataGenerator

use of io.druid.benchmark.datagen.BenchmarkDataGenerator in project druid by druid-io.

the class TopNBenchmark method setup.

@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()));
    }
    executorService = Execs.multiThreaded(numSegments, "TopNThreadPool");
    setupQueries();
    String[] schemaQuery = schemaAndQuery.split("\\.");
    String schemaName = schemaQuery[0];
    String queryName = schemaQuery[1];
    schemaInfo = BenchmarkSchemas.SCHEMA_MAP.get(schemaName);
    queryBuilder = SCHEMA_QUERY_MAP.get(schemaName).get(queryName);
    queryBuilder.threshold(threshold);
    query = queryBuilder.build();
    incIndexes = new ArrayList<>();
    for (int i = 0; i < numSegments; i++) {
        log.info("Generating rows for segment " + i);
        BenchmarkDataGenerator gen = new BenchmarkDataGenerator(schemaInfo.getColumnSchemas(), RNG_SEED + i, schemaInfo.getDataInterval(), rowsPerSegment);
        IncrementalIndex incIndex = makeIncIndex();
        for (int j = 0; j < rowsPerSegment; j++) {
            InputRow row = gen.nextRow();
            if (j % 10000 == 0) {
                log.info(j + " rows generated.");
            }
            incIndex.add(row);
        }
        incIndexes.add(incIndex);
    }
    tmpDir = Files.createTempDir();
    log.info("Using temp dir: " + tmpDir.getAbsolutePath());
    qIndexes = new ArrayList<>();
    for (int i = 0; i < numSegments; i++) {
        File indexFile = INDEX_MERGER_V9.persist(incIndexes.get(i), tmpDir, new IndexSpec());
        QueryableIndex qIndex = INDEX_IO.loadIndex(indexFile);
        qIndexes.add(qIndex);
    }
    factory = new TopNQueryRunnerFactory(new StupidPool<>("TopNBenchmark-compute-bufferPool", new OffheapBufferGenerator("compute", 250000000), 0, Integer.MAX_VALUE), new TopNQueryQueryToolChest(new TopNQueryConfig(), QueryBenchmarkUtil.NoopIntervalChunkingQueryRunnerDecorator()), QueryBenchmarkUtil.NOOP_QUERYWATCHER);
}
Also used : IndexSpec(io.druid.segment.IndexSpec) IncrementalIndex(io.druid.segment.incremental.IncrementalIndex) OnheapIncrementalIndex(io.druid.segment.incremental.OnheapIncrementalIndex) BenchmarkDataGenerator(io.druid.benchmark.datagen.BenchmarkDataGenerator) HyperUniquesSerde(io.druid.query.aggregation.hyperloglog.HyperUniquesSerde) OffheapBufferGenerator(io.druid.offheap.OffheapBufferGenerator) TopNQueryConfig(io.druid.query.topn.TopNQueryConfig) QueryableIndex(io.druid.segment.QueryableIndex) InputRow(io.druid.data.input.InputRow) TopNQueryRunnerFactory(io.druid.query.topn.TopNQueryRunnerFactory) StupidPool(io.druid.collections.StupidPool) TopNQueryQueryToolChest(io.druid.query.topn.TopNQueryQueryToolChest) File(java.io.File) Setup(org.openjdk.jmh.annotations.Setup)

Example 8 with BenchmarkDataGenerator

use of io.druid.benchmark.datagen.BenchmarkDataGenerator in project druid by druid-io.

the class BenchmarkDataGeneratorTest method testDiscreteUniform.

@Test
public void testDiscreteUniform() throws Exception {
    List<BenchmarkColumnSchema> schemas = new ArrayList<>();
    RowValueTracker tracker = new RowValueTracker();
    schemas.add(BenchmarkColumnSchema.makeDiscreteUniform("dimA", ValueType.STRING, false, 1, null, 10, 20));
    schemas.add(BenchmarkColumnSchema.makeEnumeratedDiscreteUniform("dimB", ValueType.STRING, false, 4, null, Arrays.<Object>asList("Hello", "World", "Foo", "Bar")));
    schemas.add(BenchmarkColumnSchema.makeDiscreteUniform("dimC", ValueType.STRING, false, 1, 0.50, 10, 20));
    schemas.add(BenchmarkColumnSchema.makeDiscreteUniform("dimD", ValueType.FLOAT, false, 1, null, 100, 120));
    BenchmarkDataGenerator dataGenerator = new BenchmarkDataGenerator(schemas, 9999, 0, 0, 1000.0);
    for (int i = 0; i < 100; i++) {
        InputRow row = dataGenerator.nextRow();
        //System.out.println("U-ROW: " + row);
        tracker.addRow(row);
    }
    tracker.printStuff();
}
Also used : ArrayList(java.util.ArrayList) BenchmarkDataGenerator(io.druid.benchmark.datagen.BenchmarkDataGenerator) InputRow(io.druid.data.input.InputRow) BenchmarkColumnSchema(io.druid.benchmark.datagen.BenchmarkColumnSchema) Test(org.junit.Test)

Example 9 with BenchmarkDataGenerator

use of io.druid.benchmark.datagen.BenchmarkDataGenerator in project druid by druid-io.

the class BenchmarkDataGeneratorTest method testSequential.

@Test
public void testSequential() throws Exception {
    List<BenchmarkColumnSchema> schemas = new ArrayList<>();
    RowValueTracker tracker = new RowValueTracker();
    schemas.add(BenchmarkColumnSchema.makeSequential("dimA", ValueType.STRING, false, 1, null, 10, 20));
    schemas.add(BenchmarkColumnSchema.makeEnumeratedSequential("dimB", ValueType.STRING, false, 1, null, Arrays.<Object>asList("Hello", "World", "Foo", "Bar")));
    schemas.add(BenchmarkColumnSchema.makeSequential("dimC", ValueType.STRING, false, 1, 0.50, 30, 40));
    BenchmarkDataGenerator dataGenerator = new BenchmarkDataGenerator(schemas, 9999, 0, 0, 1000.0);
    for (int i = 0; i < 100; i++) {
        InputRow row = dataGenerator.nextRow();
        //System.out.println("S-ROW: " + row);
        tracker.addRow(row);
    }
    tracker.printStuff();
}
Also used : ArrayList(java.util.ArrayList) BenchmarkDataGenerator(io.druid.benchmark.datagen.BenchmarkDataGenerator) InputRow(io.druid.data.input.InputRow) BenchmarkColumnSchema(io.druid.benchmark.datagen.BenchmarkColumnSchema) Test(org.junit.Test)

Example 10 with BenchmarkDataGenerator

use of io.druid.benchmark.datagen.BenchmarkDataGenerator in project druid by druid-io.

the class GroupByTypeInterfaceBenchmark method setup.

@Setup(Level.Trial)
public void setup() throws IOException {
    log.info("SETUP CALLED AT %d", System.currentTimeMillis());
    if (ComplexMetrics.getSerdeForType("hyperUnique") == null) {
        ComplexMetrics.registerSerde("hyperUnique", new HyperUniquesSerde(HyperLogLogHash.getDefault()));
    }
    executorService = Execs.multiThreaded(numProcessingThreads, "GroupByThreadPool[%d]");
    setupQueries();
    String schemaName = "basic";
    schemaInfo = BenchmarkSchemas.SCHEMA_MAP.get(schemaName);
    stringQuery = SCHEMA_QUERY_MAP.get(schemaName).get("string");
    longFloatQuery = SCHEMA_QUERY_MAP.get(schemaName).get("longFloat");
    longQuery = SCHEMA_QUERY_MAP.get(schemaName).get("long");
    floatQuery = SCHEMA_QUERY_MAP.get(schemaName).get("float");
    final BenchmarkDataGenerator dataGenerator = new BenchmarkDataGenerator(schemaInfo.getColumnSchemas(), RNG_SEED + 1, schemaInfo.getDataInterval(), rowsPerSegment);
    tmpDir = Files.createTempDir();
    log.info("Using temp dir: %s", tmpDir.getAbsolutePath());
    // queryableIndexes   -> numSegments worth of on-disk segments
    // anIncrementalIndex -> the last incremental index
    anIncrementalIndex = null;
    queryableIndexes = new ArrayList<>(numSegments);
    for (int i = 0; i < numSegments; i++) {
        log.info("Generating rows for segment %d/%d", i + 1, numSegments);
        final IncrementalIndex index = makeIncIndex();
        for (int j = 0; j < rowsPerSegment; j++) {
            final InputRow row = dataGenerator.nextRow();
            if (j % 20000 == 0) {
                log.info("%,d/%,d rows generated.", i * rowsPerSegment + j, rowsPerSegment * numSegments);
            }
            index.add(row);
        }
        log.info("%,d/%,d rows generated, persisting segment %d/%d.", (i + 1) * rowsPerSegment, rowsPerSegment * numSegments, i + 1, numSegments);
        final File file = INDEX_MERGER_V9.persist(index, new File(tmpDir, String.valueOf(i)), new IndexSpec());
        queryableIndexes.add(INDEX_IO.loadIndex(file));
        if (i == numSegments - 1) {
            anIncrementalIndex = index;
        } else {
            index.close();
        }
    }
    StupidPool<ByteBuffer> bufferPool = new StupidPool<>("GroupByBenchmark-computeBufferPool", new OffheapBufferGenerator("compute", 250_000_000), 0, Integer.MAX_VALUE);
    // limit of 2 is required since we simulate both historical merge and broker merge in the same process
    BlockingPool<ByteBuffer> mergePool = new BlockingPool<>(new OffheapBufferGenerator("merge", 250_000_000), 2);
    final GroupByQueryConfig config = new GroupByQueryConfig() {

        @Override
        public String getDefaultStrategy() {
            return defaultStrategy;
        }

        @Override
        public int getBufferGrouperInitialBuckets() {
            return initialBuckets;
        }

        @Override
        public long getMaxOnDiskStorage() {
            return 1_000_000_000L;
        }
    };
    config.setSingleThreaded(false);
    config.setMaxIntermediateRows(Integer.MAX_VALUE);
    config.setMaxResults(Integer.MAX_VALUE);
    DruidProcessingConfig druidProcessingConfig = new DruidProcessingConfig() {

        @Override
        public int getNumThreads() {
            // Used by "v2" strategy for concurrencyHint
            return numProcessingThreads;
        }

        @Override
        public String getFormatString() {
            return null;
        }
    };
    final Supplier<GroupByQueryConfig> configSupplier = Suppliers.ofInstance(config);
    final GroupByStrategySelector strategySelector = new GroupByStrategySelector(configSupplier, new GroupByStrategyV1(configSupplier, new GroupByQueryEngine(configSupplier, bufferPool), QueryBenchmarkUtil.NOOP_QUERYWATCHER, bufferPool), new GroupByStrategyV2(druidProcessingConfig, configSupplier, bufferPool, mergePool, new ObjectMapper(new SmileFactory()), QueryBenchmarkUtil.NOOP_QUERYWATCHER));
    factory = new GroupByQueryRunnerFactory(strategySelector, new GroupByQueryQueryToolChest(strategySelector, QueryBenchmarkUtil.NoopIntervalChunkingQueryRunnerDecorator()));
}
Also used : GroupByStrategySelector(io.druid.query.groupby.strategy.GroupByStrategySelector) IndexSpec(io.druid.segment.IndexSpec) BenchmarkDataGenerator(io.druid.benchmark.datagen.BenchmarkDataGenerator) HyperUniquesSerde(io.druid.query.aggregation.hyperloglog.HyperUniquesSerde) GroupByQueryQueryToolChest(io.druid.query.groupby.GroupByQueryQueryToolChest) GroupByStrategyV1(io.druid.query.groupby.strategy.GroupByStrategyV1) GroupByStrategyV2(io.druid.query.groupby.strategy.GroupByStrategyV2) GroupByQueryEngine(io.druid.query.groupby.GroupByQueryEngine) DefaultObjectMapper(io.druid.jackson.DefaultObjectMapper) ObjectMapper(com.fasterxml.jackson.databind.ObjectMapper) GroupByQueryRunnerFactory(io.druid.query.groupby.GroupByQueryRunnerFactory) IncrementalIndex(io.druid.segment.incremental.IncrementalIndex) OnheapIncrementalIndex(io.druid.segment.incremental.OnheapIncrementalIndex) GroupByQueryConfig(io.druid.query.groupby.GroupByQueryConfig) ByteBuffer(java.nio.ByteBuffer) SmileFactory(com.fasterxml.jackson.dataformat.smile.SmileFactory) OffheapBufferGenerator(io.druid.offheap.OffheapBufferGenerator) InputRow(io.druid.data.input.InputRow) BlockingPool(io.druid.collections.BlockingPool) StupidPool(io.druid.collections.StupidPool) DruidProcessingConfig(io.druid.query.DruidProcessingConfig) File(java.io.File) Setup(org.openjdk.jmh.annotations.Setup)

Aggregations

BenchmarkDataGenerator (io.druid.benchmark.datagen.BenchmarkDataGenerator)22 InputRow (io.druid.data.input.InputRow)22 HyperUniquesSerde (io.druid.query.aggregation.hyperloglog.HyperUniquesSerde)14 Setup (org.openjdk.jmh.annotations.Setup)14 IndexSpec (io.druid.segment.IndexSpec)10 File (java.io.File)9 ArrayList (java.util.ArrayList)9 BenchmarkColumnSchema (io.druid.benchmark.datagen.BenchmarkColumnSchema)8 IncrementalIndex (io.druid.segment.incremental.IncrementalIndex)8 OnheapIncrementalIndex (io.druid.segment.incremental.OnheapIncrementalIndex)8 Test (org.junit.Test)8 QueryableIndex (io.druid.segment.QueryableIndex)7 StupidPool (io.druid.collections.StupidPool)4 OffheapBufferGenerator (io.druid.offheap.OffheapBufferGenerator)4 Interval (org.joda.time.Interval)3 ObjectMapper (com.fasterxml.jackson.databind.ObjectMapper)2 SmileFactory (com.fasterxml.jackson.dataformat.smile.SmileFactory)2 BenchmarkSchemaInfo (io.druid.benchmark.datagen.BenchmarkSchemaInfo)2 BlockingPool (io.druid.collections.BlockingPool)2 DefaultObjectMapper (io.druid.jackson.DefaultObjectMapper)2