Examples with GroupByQueryRunnerFactory io.druid.query.groupby.GroupByQueryRunnerFactory used on opensource projects

Example 1 with GroupByQueryRunnerFactory

use of io.druid.query.groupby.GroupByQueryRunnerFactory in project druid by druid-io.

the class ApproximateHistogramGroupByQueryTest method constructorFeeder.

@Parameterized.Parameters(name = "{0}")
public static Iterable<Object[]> constructorFeeder() throws IOException {
    final GroupByQueryConfig v1Config = new GroupByQueryConfig() {

        @Override
        public String getDefaultStrategy() {
            return GroupByStrategySelector.STRATEGY_V1;
        }

        @Override
        public String toString() {
            return "v1";
        }
    };
    final GroupByQueryConfig v1SingleThreadedConfig = new GroupByQueryConfig() {

        @Override
        public boolean isSingleThreaded() {
            return true;
        }

        @Override
        public String getDefaultStrategy() {
            return GroupByStrategySelector.STRATEGY_V1;
        }

        @Override
        public String toString() {
            return "v1SingleThreaded";
        }
    };
    final GroupByQueryConfig v2Config = new GroupByQueryConfig() {

        @Override
        public String getDefaultStrategy() {
            return GroupByStrategySelector.STRATEGY_V2;
        }

        @Override
        public String toString() {
            return "v2";
        }
    };
    v1Config.setMaxIntermediateRows(10000);
    v1SingleThreadedConfig.setMaxIntermediateRows(10000);
    final List<Object[]> constructors = Lists.newArrayList();
    final List<GroupByQueryConfig> configs = ImmutableList.of(v1Config, v1SingleThreadedConfig, v2Config);
    for (GroupByQueryConfig config : configs) {
        final GroupByQueryRunnerFactory factory = GroupByQueryRunnerTest.makeQueryRunnerFactory(config);
        for (QueryRunner<Row> runner : QueryRunnerTestHelper.makeQueryRunners(factory)) {
            final String testName = String.format("config=%s, runner=%s", config.toString(), runner.toString());
            constructors.add(new Object[] { testName, factory, runner });
        }
    }
    return constructors;
}

Example 2 with GroupByQueryRunnerFactory

use of io.druid.query.groupby.GroupByQueryRunnerFactory in project druid by druid-io.

the class AggregationTestHelper method createGroupByQueryAggregationTestHelper.

public static final AggregationTestHelper createGroupByQueryAggregationTestHelper(List<? extends Module> jsonModulesToRegister, GroupByQueryConfig config, TemporaryFolder tempFolder) {
    ObjectMapper mapper = new DefaultObjectMapper();
    GroupByQueryRunnerFactory factory = GroupByQueryRunnerTest.makeQueryRunnerFactory(mapper, config);
    IndexIO indexIO = new IndexIO(mapper, new ColumnConfig() {

        @Override
        public int columnCacheSizeBytes() {
            return 0;
        }
    });
    return new AggregationTestHelper(mapper, new IndexMerger(mapper, indexIO), indexIO, factory.getToolchest(), factory, tempFolder, jsonModulesToRegister);
}

Example 3 with GroupByQueryRunnerFactory

use of io.druid.query.groupby.GroupByQueryRunnerFactory 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()));
}

Example 4 with GroupByQueryRunnerFactory

use of io.druid.query.groupby.GroupByQueryRunnerFactory in project druid by druid-io.

the class GroupByBenchmark method setup.

@Setup(Level.Trial)
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(numProcessingThreads, "GroupByThreadPool[%d]");
    setupQueries();
    String[] schemaQuery = schemaAndQuery.split("\\.");
    String schemaName = schemaQuery[0];
    String queryName = schemaQuery[1];
    schemaInfo = BenchmarkSchemas.SCHEMA_MAP.get(schemaName);
    query = SCHEMA_QUERY_MAP.get(schemaName).get(queryName);
    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(schemaInfo.isWithRollup());
        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()));
}

Example 5 with GroupByQueryRunnerFactory

use of io.druid.query.groupby.GroupByQueryRunnerFactory in project druid by druid-io.

the class DistinctCountGroupByQueryTest method testGroupByWithDistinctCountAgg.

@Test
public void testGroupByWithDistinctCountAgg() throws Exception {
    final GroupByQueryConfig config = new GroupByQueryConfig();
    config.setMaxIntermediateRows(10000);
    final GroupByQueryRunnerFactory factory = GroupByQueryRunnerTest.makeQueryRunnerFactory(config);
    IncrementalIndex index = new OnheapIncrementalIndex(0, Granularities.SECOND, new AggregatorFactory[] { new CountAggregatorFactory("cnt") }, 1000);
    String visitor_id = "visitor_id";
    String client_type = "client_type";
    long timestamp = System.currentTimeMillis();
    index.add(new MapBasedInputRow(timestamp, Lists.newArrayList(visitor_id, client_type), ImmutableMap.<String, Object>of(visitor_id, "0", client_type, "iphone")));
    index.add(new MapBasedInputRow(timestamp + 1, Lists.newArrayList(visitor_id, client_type), ImmutableMap.<String, Object>of(visitor_id, "1", client_type, "iphone")));
    index.add(new MapBasedInputRow(timestamp + 2, Lists.newArrayList(visitor_id, client_type), ImmutableMap.<String, Object>of(visitor_id, "2", client_type, "android")));
    GroupByQuery query = new GroupByQuery.Builder().setDataSource(QueryRunnerTestHelper.dataSource).setGranularity(QueryRunnerTestHelper.allGran).setDimensions(Arrays.<DimensionSpec>asList(new DefaultDimensionSpec(client_type, client_type))).setInterval(QueryRunnerTestHelper.fullOnInterval).setLimitSpec(new DefaultLimitSpec(Lists.newArrayList(new OrderByColumnSpec(client_type, OrderByColumnSpec.Direction.DESCENDING)), 10)).setAggregatorSpecs(Lists.newArrayList(QueryRunnerTestHelper.rowsCount, new DistinctCountAggregatorFactory("UV", visitor_id, null))).build();
    final Segment incrementalIndexSegment = new IncrementalIndexSegment(index, null);
    Iterable<Row> results = GroupByQueryRunnerTestHelper.runQuery(factory, factory.createRunner(incrementalIndexSegment), query);
    List<Row> expectedResults = Arrays.asList(GroupByQueryRunnerTestHelper.createExpectedRow("1970-01-01T00:00:00.000Z", client_type, "iphone", "UV", 2L, "rows", 2L), GroupByQueryRunnerTestHelper.createExpectedRow("1970-01-01T00:00:00.000Z", client_type, "android", "UV", 1L, "rows", 1L));
    TestHelper.assertExpectedObjects(expectedResults, results, "distinct-count");
}