Examples with SpecificSegmentQueryRunner org.apache.druid.query.spec.SpecificSegmentQueryRunner used on opensource projects

Example 1 with SpecificSegmentQueryRunner

use of org.apache.druid.query.spec.SpecificSegmentQueryRunner in project druid by druid-io.

the class ServerManager method buildAndDecorateQueryRunner.

private <T> QueryRunner<T> buildAndDecorateQueryRunner(final QueryRunnerFactory<T, Query<T>> factory, final QueryToolChest<T, Query<T>> toolChest, final SegmentReference segment, final Optional<byte[]> cacheKeyPrefix, final SegmentDescriptor segmentDescriptor, final AtomicLong cpuTimeAccumulator) {
    final SpecificSegmentSpec segmentSpec = new SpecificSegmentSpec(segmentDescriptor);
    final SegmentId segmentId = segment.getId();
    final Interval segmentInterval = segment.getDataInterval();
    // If the segment is closed after this line, ReferenceCountingSegmentQueryRunner will handle and do the right thing.
    if (segmentId == null || segmentInterval == null) {
        return new ReportTimelineMissingSegmentQueryRunner<>(segmentDescriptor);
    }
    String segmentIdString = segmentId.toString();
    MetricsEmittingQueryRunner<T> metricsEmittingQueryRunnerInner = new MetricsEmittingQueryRunner<>(emitter, toolChest, new ReferenceCountingSegmentQueryRunner<>(factory, segment, segmentDescriptor), QueryMetrics::reportSegmentTime, queryMetrics -> queryMetrics.segment(segmentIdString));
    StorageAdapter storageAdapter = segment.asStorageAdapter();
    long segmentMaxTime = storageAdapter.getMaxTime().getMillis();
    long segmentMinTime = storageAdapter.getMinTime().getMillis();
    Interval actualDataInterval = Intervals.utc(segmentMinTime, segmentMaxTime + 1);
    CachingQueryRunner<T> cachingQueryRunner = new CachingQueryRunner<>(segmentIdString, cacheKeyPrefix, segmentDescriptor, actualDataInterval, objectMapper, cache, toolChest, metricsEmittingQueryRunnerInner, cachePopulator, cacheConfig);
    BySegmentQueryRunner<T> bySegmentQueryRunner = new BySegmentQueryRunner<>(segmentId, segmentInterval.getStart(), cachingQueryRunner);
    MetricsEmittingQueryRunner<T> metricsEmittingQueryRunnerOuter = new MetricsEmittingQueryRunner<>(emitter, toolChest, bySegmentQueryRunner, QueryMetrics::reportSegmentAndCacheTime, queryMetrics -> queryMetrics.segment(segmentIdString)).withWaitMeasuredFromNow();
    SpecificSegmentQueryRunner<T> specificSegmentQueryRunner = new SpecificSegmentQueryRunner<>(metricsEmittingQueryRunnerOuter, segmentSpec);
    PerSegmentOptimizingQueryRunner<T> perSegmentOptimizingQueryRunner = new PerSegmentOptimizingQueryRunner<>(specificSegmentQueryRunner, new PerSegmentQueryOptimizationContext(segmentDescriptor));
    return new SetAndVerifyContextQueryRunner<>(serverConfig, CPUTimeMetricQueryRunner.safeBuild(perSegmentOptimizingQueryRunner, toolChest, emitter, cpuTimeAccumulator, false));
}

Example 2 with SpecificSegmentQueryRunner

use of org.apache.druid.query.spec.SpecificSegmentQueryRunner in project druid by druid-io.

the class SinkQuerySegmentWalker method getQueryRunnerForSegments.

@Override
public <T> QueryRunner<T> getQueryRunnerForSegments(final Query<T> query, final Iterable<SegmentDescriptor> specs) {
    // We only handle one particular dataSource. Make sure that's what we have, then ignore from here on out.
    final DataSourceAnalysis analysis = DataSourceAnalysis.forDataSource(query.getDataSource());
    // Sanity check: make sure the query is based on the table we're meant to handle.
    if (!analysis.getBaseTableDataSource().filter(ds -> dataSource.equals(ds.getName())).isPresent()) {
        throw new ISE("Cannot handle datasource: %s", analysis.getDataSource());
    }
    final QueryRunnerFactory<T, Query<T>> factory = conglomerate.findFactory(query);
    if (factory == null) {
        throw new ISE("Unknown query type[%s].", query.getClass());
    }
    final QueryToolChest<T, Query<T>> toolChest = factory.getToolchest();
    final boolean skipIncrementalSegment = query.getContextValue(CONTEXT_SKIP_INCREMENTAL_SEGMENT, false);
    final AtomicLong cpuTimeAccumulator = new AtomicLong(0L);
    // Make sure this query type can handle the subquery, if present.
    if (analysis.isQuery() && !toolChest.canPerformSubquery(((QueryDataSource) analysis.getDataSource()).getQuery())) {
        throw new ISE("Cannot handle subquery: %s", analysis.getDataSource());
    }
    // segmentMapFn maps each base Segment into a joined Segment if necessary.
    final Function<SegmentReference, SegmentReference> segmentMapFn = joinableFactoryWrapper.createSegmentMapFn(analysis.getJoinBaseTableFilter().map(Filters::toFilter).orElse(null), analysis.getPreJoinableClauses(), cpuTimeAccumulator, analysis.getBaseQuery().orElse(query));
    // We compute the join cache key here itself so it doesn't need to be re-computed for every segment
    final Optional<byte[]> cacheKeyPrefix = analysis.isJoin() ? joinableFactoryWrapper.computeJoinDataSourceCacheKey(analysis) : Optional.of(StringUtils.EMPTY_BYTES);
    Iterable<QueryRunner<T>> perSegmentRunners = Iterables.transform(specs, descriptor -> {
        final PartitionChunk<Sink> chunk = sinkTimeline.findChunk(descriptor.getInterval(), descriptor.getVersion(), descriptor.getPartitionNumber());
        if (chunk == null) {
            return new ReportTimelineMissingSegmentQueryRunner<>(descriptor);
        }
        final Sink theSink = chunk.getObject();
        final SegmentId sinkSegmentId = theSink.getSegment().getId();
        Iterable<QueryRunner<T>> perHydrantRunners = new SinkQueryRunners<>(Iterables.transform(theSink, hydrant -> {
            // Hydrant might swap at any point, but if it's swapped at the start
            // then we know it's *definitely* swapped.
            final boolean hydrantDefinitelySwapped = hydrant.hasSwapped();
            if (skipIncrementalSegment && !hydrantDefinitelySwapped) {
                return new Pair<>(hydrant.getSegmentDataInterval(), new NoopQueryRunner<>());
            }
            // Prevent the underlying segment from swapping when its being iterated
            final Optional<Pair<SegmentReference, Closeable>> maybeSegmentAndCloseable = hydrant.getSegmentForQuery(segmentMapFn);
            // if optional isn't present, we failed to acquire reference to the segment or any joinables
            if (!maybeSegmentAndCloseable.isPresent()) {
                return new Pair<>(hydrant.getSegmentDataInterval(), new ReportTimelineMissingSegmentQueryRunner<>(descriptor));
            }
            final Pair<SegmentReference, Closeable> segmentAndCloseable = maybeSegmentAndCloseable.get();
            try {
                QueryRunner<T> runner = factory.createRunner(segmentAndCloseable.lhs);
                // 2) Hydrants are not the same between replicas, make sure cache is local
                if (hydrantDefinitelySwapped && cache.isLocal()) {
                    StorageAdapter storageAdapter = segmentAndCloseable.lhs.asStorageAdapter();
                    long segmentMinTime = storageAdapter.getMinTime().getMillis();
                    long segmentMaxTime = storageAdapter.getMaxTime().getMillis();
                    Interval actualDataInterval = Intervals.utc(segmentMinTime, segmentMaxTime + 1);
                    runner = new CachingQueryRunner<>(makeHydrantCacheIdentifier(hydrant), cacheKeyPrefix, descriptor, actualDataInterval, objectMapper, cache, toolChest, runner, // Always populate in foreground regardless of config
                    new ForegroundCachePopulator(objectMapper, cachePopulatorStats, cacheConfig.getMaxEntrySize()), cacheConfig);
                }
                // Make it always use Closeable to decrement()
                runner = QueryRunnerHelper.makeClosingQueryRunner(runner, segmentAndCloseable.rhs);
                return new Pair<>(segmentAndCloseable.lhs.getDataInterval(), runner);
            } catch (Throwable e) {
                throw CloseableUtils.closeAndWrapInCatch(e, segmentAndCloseable.rhs);
            }
        }));
        return new SpecificSegmentQueryRunner<>(withPerSinkMetrics(new BySegmentQueryRunner<>(sinkSegmentId, descriptor.getInterval().getStart(), factory.mergeRunners(DirectQueryProcessingPool.INSTANCE, perHydrantRunners)), toolChest, sinkSegmentId, cpuTimeAccumulator), new SpecificSegmentSpec(descriptor));
    });
    final QueryRunner<T> mergedRunner = toolChest.mergeResults(factory.mergeRunners(queryProcessingPool, perSegmentRunners));
    return CPUTimeMetricQueryRunner.safeBuild(new FinalizeResultsQueryRunner<>(mergedRunner, toolChest), toolChest, emitter, cpuTimeAccumulator, true);
}