Examples with BenchmarkSchemaInfo io.druid.benchmark.datagen.BenchmarkSchemaInfo used on opensource projects

Example 6 with BenchmarkSchemaInfo

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

the class TopNBenchmark method setupQueries.

private void setupQueries() {
    // queries for the basic schema
    Map<String, TopNQueryBuilder> basicQueries = new LinkedHashMap<>();
    BenchmarkSchemaInfo basicSchema = BenchmarkSchemas.SCHEMA_MAP.get("basic");
    {
        // basic.A
        QuerySegmentSpec intervalSpec = new MultipleIntervalSegmentSpec(Arrays.asList(basicSchema.getDataInterval()));
        List<AggregatorFactory> queryAggs = new ArrayList<>();
        queryAggs.add(new LongSumAggregatorFactory("sumLongSequential", "sumLongSequential"));
        queryAggs.add(new LongMaxAggregatorFactory("maxLongUniform", "maxLongUniform"));
        queryAggs.add(new DoubleSumAggregatorFactory("sumFloatNormal", "sumFloatNormal"));
        queryAggs.add(new DoubleMinAggregatorFactory("minFloatZipf", "minFloatZipf"));
        queryAggs.add(new HyperUniquesAggregatorFactory("hyperUniquesMet", "hyper"));
        TopNQueryBuilder queryBuilderA = new TopNQueryBuilder().dataSource("blah").granularity(Granularities.ALL).dimension("dimSequential").metric("sumFloatNormal").intervals(intervalSpec).aggregators(queryAggs);
        basicQueries.put("A", queryBuilderA);
    }
    {
        // basic.numericSort
        QuerySegmentSpec intervalSpec = new MultipleIntervalSegmentSpec(Arrays.asList(basicSchema.getDataInterval()));
        List<AggregatorFactory> queryAggs = new ArrayList<>();
        queryAggs.add(new LongSumAggregatorFactory("sumLongSequential", "sumLongSequential"));
        TopNQueryBuilder queryBuilderA = new TopNQueryBuilder().dataSource("blah").granularity(Granularities.ALL).dimension("dimUniform").metric(new DimensionTopNMetricSpec(null, StringComparators.NUMERIC)).intervals(intervalSpec).aggregators(queryAggs);
        basicQueries.put("numericSort", queryBuilderA);
    }
    {
        // basic.alphanumericSort
        QuerySegmentSpec intervalSpec = new MultipleIntervalSegmentSpec(Arrays.asList(basicSchema.getDataInterval()));
        List<AggregatorFactory> queryAggs = new ArrayList<>();
        queryAggs.add(new LongSumAggregatorFactory("sumLongSequential", "sumLongSequential"));
        TopNQueryBuilder queryBuilderA = new TopNQueryBuilder().dataSource("blah").granularity(Granularities.ALL).dimension("dimUniform").metric(new DimensionTopNMetricSpec(null, StringComparators.ALPHANUMERIC)).intervals(intervalSpec).aggregators(queryAggs);
        basicQueries.put("alphanumericSort", queryBuilderA);
    }
    SCHEMA_QUERY_MAP.put("basic", basicQueries);
}

Example 7 with BenchmarkSchemaInfo

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

the class FilteredAggregatorBenchmark 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()));
    }
    schemaInfo = BenchmarkSchemas.SCHEMA_MAP.get(schema);
    BenchmarkDataGenerator gen = new BenchmarkDataGenerator(schemaInfo.getColumnSchemas(), RNG_SEED, schemaInfo.getDataInterval(), rowsPerSegment);
    incIndex = makeIncIndex(schemaInfo.getAggsArray());
    filter = new OrDimFilter(Arrays.asList(new BoundDimFilter("dimSequential", "-1", "-1", true, true, null, null, StringComparators.ALPHANUMERIC), new JavaScriptDimFilter("dimSequential", "function(x) { return false }", null, JavaScriptConfig.getEnabledInstance()), new RegexDimFilter("dimSequential", "X", null), new SearchQueryDimFilter("dimSequential", new ContainsSearchQuerySpec("X", false), null), new InDimFilter("dimSequential", Arrays.asList("X"), null)));
    filteredMetrics = new AggregatorFactory[1];
    filteredMetrics[0] = new FilteredAggregatorFactory(new CountAggregatorFactory("rows"), filter);
    incIndexFilteredAgg = makeIncIndex(filteredMetrics);
    inputRows = new ArrayList<>();
    for (int j = 0; j < rowsPerSegment; j++) {
        InputRow row = gen.nextRow();
        if (j % 10000 == 0) {
            log.info(j + " rows generated.");
        }
        incIndex.add(row);
        inputRows.add(row);
    }
    tmpDir = Files.createTempDir();
    log.info("Using temp dir: " + tmpDir.getAbsolutePath());
    indexFile = INDEX_MERGER_V9.persist(incIndex, tmpDir, new IndexSpec());
    qIndex = INDEX_IO.loadIndex(indexFile);
    factory = new TimeseriesQueryRunnerFactory(new TimeseriesQueryQueryToolChest(QueryBenchmarkUtil.NoopIntervalChunkingQueryRunnerDecorator()), new TimeseriesQueryEngine(), QueryBenchmarkUtil.NOOP_QUERYWATCHER);
    BenchmarkSchemaInfo basicSchema = BenchmarkSchemas.SCHEMA_MAP.get("basic");
    QuerySegmentSpec intervalSpec = new MultipleIntervalSegmentSpec(Arrays.asList(basicSchema.getDataInterval()));
    List<AggregatorFactory> queryAggs = new ArrayList<>();
    queryAggs.add(filteredMetrics[0]);
    query = Druids.newTimeseriesQueryBuilder().dataSource("blah").granularity(Granularities.ALL).intervals(intervalSpec).aggregators(queryAggs).descending(false).build();
}

Example 8 with BenchmarkSchemaInfo

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

the class TopNTypeInterfaceBenchmark method setupQueries.

private void setupQueries() {
    // queries for the basic schema
    Map<String, TopNQueryBuilder> basicQueries = new LinkedHashMap<>();
    BenchmarkSchemaInfo basicSchema = BenchmarkSchemas.SCHEMA_MAP.get("basic");
    {
        // basic.A
        QuerySegmentSpec intervalSpec = new MultipleIntervalSegmentSpec(Arrays.asList(basicSchema.getDataInterval()));
        List<AggregatorFactory> queryAggs = new ArrayList<>();
        queryAggs.add(new LongSumAggregatorFactory("sumLongSequential", "sumLongSequential"));
        queryAggs.add(new LongMaxAggregatorFactory("maxLongUniform", "maxLongUniform"));
        queryAggs.add(new DoubleSumAggregatorFactory("sumFloatNormal", "sumFloatNormal"));
        queryAggs.add(new DoubleMinAggregatorFactory("minFloatZipf", "minFloatZipf"));
        queryAggs.add(new HyperUniquesAggregatorFactory("hyperUniquesMet", "hyper"));
        // Use an IdentityExtractionFn to force usage of DimExtractionTopNAlgorithm
        TopNQueryBuilder queryBuilderString = new TopNQueryBuilder().dataSource("blah").granularity(Granularities.ALL).dimension(new ExtractionDimensionSpec("dimSequential", "dimSequential", IdentityExtractionFn.getInstance())).metric("sumFloatNormal").intervals(intervalSpec).aggregators(queryAggs);
        // DimExtractionTopNAlgorithm is always used for numeric columns
        TopNQueryBuilder queryBuilderLong = new TopNQueryBuilder().dataSource("blah").granularity(Granularities.ALL).dimension("metLongUniform").metric("sumFloatNormal").intervals(intervalSpec).aggregators(queryAggs);
        TopNQueryBuilder queryBuilderFloat = new TopNQueryBuilder().dataSource("blah").granularity(Granularities.ALL).dimension("metFloatNormal").metric("sumFloatNormal").intervals(intervalSpec).aggregators(queryAggs);
        basicQueries.put("string", queryBuilderString);
        basicQueries.put("long", queryBuilderLong);
        basicQueries.put("float", queryBuilderFloat);
    }
    {
        // basic.numericSort
        QuerySegmentSpec intervalSpec = new MultipleIntervalSegmentSpec(Arrays.asList(basicSchema.getDataInterval()));
        List<AggregatorFactory> queryAggs = new ArrayList<>();
        queryAggs.add(new LongSumAggregatorFactory("sumLongSequential", "sumLongSequential"));
        TopNQueryBuilder queryBuilderA = new TopNQueryBuilder().dataSource("blah").granularity(Granularities.ALL).dimension("dimUniform").metric(new DimensionTopNMetricSpec(null, StringComparators.NUMERIC)).intervals(intervalSpec).aggregators(queryAggs);
        basicQueries.put("numericSort", queryBuilderA);
    }
    {
        // basic.alphanumericSort
        QuerySegmentSpec intervalSpec = new MultipleIntervalSegmentSpec(Arrays.asList(basicSchema.getDataInterval()));
        List<AggregatorFactory> queryAggs = new ArrayList<>();
        queryAggs.add(new LongSumAggregatorFactory("sumLongSequential", "sumLongSequential"));
        TopNQueryBuilder queryBuilderA = new TopNQueryBuilder().dataSource("blah").granularity(Granularities.ALL).dimension("dimUniform").metric(new DimensionTopNMetricSpec(null, StringComparators.ALPHANUMERIC)).intervals(intervalSpec).aggregators(queryAggs);
        basicQueries.put("alphanumericSort", queryBuilderA);
    }
    SCHEMA_QUERY_MAP.put("basic", basicQueries);
}

Example 9 with BenchmarkSchemaInfo

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

the class GroupByBenchmark method setupQueries.

private void setupQueries() {
    // queries for the basic schema
    Map<String, GroupByQuery> basicQueries = new LinkedHashMap<>();
    BenchmarkSchemaInfo basicSchema = BenchmarkSchemas.SCHEMA_MAP.get("basic");
    {
        // basic.A
        QuerySegmentSpec intervalSpec = new MultipleIntervalSegmentSpec(Arrays.asList(basicSchema.getDataInterval()));
        List<AggregatorFactory> queryAggs = new ArrayList<>();
        queryAggs.add(new LongSumAggregatorFactory("sumLongSequential", "sumLongSequential"));
        GroupByQuery queryA = GroupByQuery.builder().setDataSource("blah").setQuerySegmentSpec(intervalSpec).setDimensions(Lists.<DimensionSpec>newArrayList(new DefaultDimensionSpec("dimSequential", null), new DefaultDimensionSpec("dimZipf", null))).setAggregatorSpecs(queryAggs).setGranularity(Granularity.fromString(queryGranularity)).build();
        basicQueries.put("A", queryA);
    }
    {
        // basic.nested
        QuerySegmentSpec intervalSpec = new MultipleIntervalSegmentSpec(Arrays.asList(basicSchema.getDataInterval()));
        List<AggregatorFactory> queryAggs = new ArrayList<>();
        queryAggs.add(new LongSumAggregatorFactory("sumLongSequential", "sumLongSequential"));
        GroupByQuery subqueryA = GroupByQuery.builder().setDataSource("blah").setQuerySegmentSpec(intervalSpec).setDimensions(Lists.<DimensionSpec>newArrayList(new DefaultDimensionSpec("dimSequential", null), new DefaultDimensionSpec("dimZipf", null))).setAggregatorSpecs(queryAggs).setGranularity(Granularities.DAY).build();
        GroupByQuery queryA = GroupByQuery.builder().setDataSource(subqueryA).setQuerySegmentSpec(intervalSpec).setDimensions(Lists.<DimensionSpec>newArrayList(new DefaultDimensionSpec("dimSequential", null))).setAggregatorSpecs(queryAggs).setGranularity(Granularities.WEEK).build();
        basicQueries.put("nested", queryA);
    }
    SCHEMA_QUERY_MAP.put("basic", basicQueries);
    // simple one column schema, for testing performance difference between querying on numeric values as Strings and
    // directly as longs
    Map<String, GroupByQuery> simpleQueries = new LinkedHashMap<>();
    BenchmarkSchemaInfo simpleSchema = BenchmarkSchemas.SCHEMA_MAP.get("simple");
    {
        // simple.A
        QuerySegmentSpec intervalSpec = new MultipleIntervalSegmentSpec(Arrays.asList(simpleSchema.getDataInterval()));
        List<AggregatorFactory> queryAggs = new ArrayList<>();
        queryAggs.add(new LongSumAggregatorFactory("rows", "rows"));
        GroupByQuery queryA = GroupByQuery.builder().setDataSource("blah").setQuerySegmentSpec(intervalSpec).setDimensions(Lists.<DimensionSpec>newArrayList(new DefaultDimensionSpec("dimSequential", "dimSequential", ValueType.STRING))).setAggregatorSpecs(queryAggs).setGranularity(Granularity.fromString(queryGranularity)).build();
        simpleQueries.put("A", queryA);
    }
    SCHEMA_QUERY_MAP.put("simple", simpleQueries);
    Map<String, GroupByQuery> simpleLongQueries = new LinkedHashMap<>();
    BenchmarkSchemaInfo simpleLongSchema = BenchmarkSchemas.SCHEMA_MAP.get("simpleLong");
    {
        // simpleLong.A
        QuerySegmentSpec intervalSpec = new MultipleIntervalSegmentSpec(Arrays.asList(simpleLongSchema.getDataInterval()));
        List<AggregatorFactory> queryAggs = new ArrayList<>();
        queryAggs.add(new LongSumAggregatorFactory("rows", "rows"));
        GroupByQuery queryA = GroupByQuery.builder().setDataSource("blah").setQuerySegmentSpec(intervalSpec).setDimensions(Lists.<DimensionSpec>newArrayList(new DefaultDimensionSpec("dimSequential", "dimSequential", ValueType.LONG))).setAggregatorSpecs(queryAggs).setGranularity(Granularity.fromString(queryGranularity)).build();
        simpleLongQueries.put("A", queryA);
    }
    SCHEMA_QUERY_MAP.put("simpleLong", simpleLongQueries);
    Map<String, GroupByQuery> simpleFloatQueries = new LinkedHashMap<>();
    BenchmarkSchemaInfo simpleFloatSchema = BenchmarkSchemas.SCHEMA_MAP.get("simpleFloat");
    {
        // simpleFloat.A
        QuerySegmentSpec intervalSpec = new MultipleIntervalSegmentSpec(Arrays.asList(simpleFloatSchema.getDataInterval()));
        List<AggregatorFactory> queryAggs = new ArrayList<>();
        queryAggs.add(new LongSumAggregatorFactory("rows", "rows"));
        GroupByQuery queryA = GroupByQuery.builder().setDataSource("blah").setQuerySegmentSpec(intervalSpec).setDimensions(Lists.<DimensionSpec>newArrayList(new DefaultDimensionSpec("dimSequential", "dimSequential", ValueType.FLOAT))).setAggregatorSpecs(queryAggs).setGranularity(Granularity.fromString(queryGranularity)).build();
        simpleFloatQueries.put("A", queryA);
    }
    SCHEMA_QUERY_MAP.put("simpleFloat", simpleFloatQueries);
}