Examples with SpillSpaceTracker com.facebook.presto.spiller.SpillSpaceTracker used on opensource projects

Example 1 with SpillSpaceTracker

use of com.facebook.presto.spiller.SpillSpaceTracker in project presto by prestodb.

the class PrestoSparkTaskExecutorFactory method doCreate.

public <T extends PrestoSparkTaskOutput> IPrestoSparkTaskExecutor<T> doCreate(int partitionId, int attemptNumber, SerializedPrestoSparkTaskDescriptor serializedTaskDescriptor, Iterator<SerializedPrestoSparkTaskSource> serializedTaskSources, PrestoSparkTaskInputs inputs, CollectionAccumulator<SerializedTaskInfo> taskInfoCollector, CollectionAccumulator<PrestoSparkShuffleStats> shuffleStatsCollector, Class<T> outputType) {
    PrestoSparkTaskDescriptor taskDescriptor = taskDescriptorJsonCodec.fromJson(serializedTaskDescriptor.getBytes());
    ImmutableMap.Builder<String, TokenAuthenticator> extraAuthenticators = ImmutableMap.builder();
    authenticatorProviders.forEach(provider -> extraAuthenticators.putAll(provider.getTokenAuthenticators()));
    Session session = taskDescriptor.getSession().toSession(sessionPropertyManager, taskDescriptor.getExtraCredentials(), extraAuthenticators.build());
    PlanFragment fragment = taskDescriptor.getFragment();
    StageId stageId = new StageId(session.getQueryId(), fragment.getId().getId());
    // Clear the cache if the cache does not have broadcast table for current stageId.
    // We will only cache 1 HT at any time. If the stageId changes, we will drop the old cached HT
    prestoSparkBroadcastTableCacheManager.removeCachedTablesForStagesOtherThan(stageId);
    // TODO: include attemptId in taskId
    TaskId taskId = new TaskId(new StageExecutionId(stageId, 0), partitionId);
    List<TaskSource> taskSources = getTaskSources(serializedTaskSources);
    log.info("Task [%s] received %d splits.", taskId, taskSources.stream().mapToInt(taskSource -> taskSource.getSplits().size()).sum());
    OptionalLong totalSplitSize = computeAllSplitsSize(taskSources);
    if (totalSplitSize.isPresent()) {
        log.info("Total split size: %s bytes.", totalSplitSize.getAsLong());
    }
    // TODO: Remove this once we can display the plan on Spark UI.
    log.info(PlanPrinter.textPlanFragment(fragment, functionAndTypeManager, session, true));
    DataSize maxUserMemory = new DataSize(min(nodeMemoryConfig.getMaxQueryMemoryPerNode().toBytes(), getQueryMaxMemoryPerNode(session).toBytes()), BYTE);
    DataSize maxTotalMemory = new DataSize(min(nodeMemoryConfig.getMaxQueryTotalMemoryPerNode().toBytes(), getQueryMaxTotalMemoryPerNode(session).toBytes()), BYTE);
    DataSize maxBroadcastMemory = getSparkBroadcastJoinMaxMemoryOverride(session);
    if (maxBroadcastMemory == null) {
        maxBroadcastMemory = new DataSize(min(nodeMemoryConfig.getMaxQueryBroadcastMemory().toBytes(), getQueryMaxBroadcastMemory(session).toBytes()), BYTE);
    }
    MemoryPool memoryPool = new MemoryPool(new MemoryPoolId("spark-executor-memory-pool"), maxTotalMemory);
    SpillSpaceTracker spillSpaceTracker = new SpillSpaceTracker(maxQuerySpillPerNode);
    QueryContext queryContext = new QueryContext(session.getQueryId(), maxUserMemory, maxTotalMemory, maxBroadcastMemory, maxRevocableMemory, memoryPool, new TestingGcMonitor(), notificationExecutor, yieldExecutor, maxQuerySpillPerNode, spillSpaceTracker, memoryReservationSummaryJsonCodec);
    queryContext.setVerboseExceededMemoryLimitErrorsEnabled(isVerboseExceededMemoryLimitErrorsEnabled(session));
    queryContext.setHeapDumpOnExceededMemoryLimitEnabled(isHeapDumpOnExceededMemoryLimitEnabled(session));
    String heapDumpFilePath = Paths.get(getHeapDumpFileDirectory(session), format("%s_%s.hprof", session.getQueryId().getId(), stageId.getId())).toString();
    queryContext.setHeapDumpFilePath(heapDumpFilePath);
    TaskStateMachine taskStateMachine = new TaskStateMachine(taskId, notificationExecutor);
    TaskContext taskContext = queryContext.addTaskContext(taskStateMachine, session, // Plan has to be retained only if verbose memory exceeded errors are requested
    isVerboseExceededMemoryLimitErrorsEnabled(session) ? Optional.of(fragment.getRoot()) : Optional.empty(), perOperatorCpuTimerEnabled, cpuTimerEnabled, perOperatorAllocationTrackingEnabled, allocationTrackingEnabled, false);
    final double memoryRevokingThreshold = getMemoryRevokingThreshold(session);
    final double memoryRevokingTarget = getMemoryRevokingTarget(session);
    checkArgument(memoryRevokingTarget <= memoryRevokingThreshold, "memoryRevokingTarget should be less than or equal memoryRevokingThreshold, but got %s and %s respectively", memoryRevokingTarget, memoryRevokingThreshold);
    if (isSpillEnabled(session)) {
        memoryPool.addListener((pool, queryId, totalMemoryReservationBytes) -> {
            if (totalMemoryReservationBytes > queryContext.getPeakNodeTotalMemory()) {
                queryContext.setPeakNodeTotalMemory(totalMemoryReservationBytes);
            }
            if (totalMemoryReservationBytes > pool.getMaxBytes() * memoryRevokingThreshold && memoryRevokeRequestInProgress.compareAndSet(false, true)) {
                memoryRevocationExecutor.execute(() -> {
                    try {
                        AtomicLong remainingBytesToRevoke = new AtomicLong(totalMemoryReservationBytes - (long) (memoryRevokingTarget * pool.getMaxBytes()));
                        remainingBytesToRevoke.addAndGet(-MemoryRevokingSchedulerUtils.getMemoryAlreadyBeingRevoked(ImmutableList.of(taskContext), remainingBytesToRevoke.get()));
                        taskContext.accept(new VoidTraversingQueryContextVisitor<AtomicLong>() {

                            @Override
                            public Void visitOperatorContext(OperatorContext operatorContext, AtomicLong remainingBytesToRevoke) {
                                if (remainingBytesToRevoke.get() > 0) {
                                    long revokedBytes = operatorContext.requestMemoryRevoking();
                                    if (revokedBytes > 0) {
                                        memoryRevokePending.set(true);
                                        remainingBytesToRevoke.addAndGet(-revokedBytes);
                                    }
                                }
                                return null;
                            }
                        }, remainingBytesToRevoke);
                        memoryRevokeRequestInProgress.set(false);
                    } catch (Exception e) {
                        log.error(e, "Error requesting memory revoking");
                    }
                });
            }
            // Get the latest memory reservation info since it might have changed due to revoke
            long totalReservedMemory = pool.getQueryMemoryReservation(queryId) + pool.getQueryRevocableMemoryReservation(queryId);
            // If total memory usage is over maxTotalMemory and memory revoke request is not pending, fail the query with EXCEEDED_MEMORY_LIMIT error
            if (totalReservedMemory > maxTotalMemory.toBytes() && !memoryRevokeRequestInProgress.get() && !isMemoryRevokePending(taskContext)) {
                throw exceededLocalTotalMemoryLimit(maxTotalMemory, queryContext.getAdditionalFailureInfo(totalReservedMemory, 0) + format("Total reserved memory: %s, Total revocable memory: %s", succinctBytes(pool.getQueryMemoryReservation(queryId)), succinctBytes(pool.getQueryRevocableMemoryReservation(queryId))), isHeapDumpOnExceededMemoryLimitEnabled(session), Optional.ofNullable(heapDumpFilePath));
            }
        });
    }
    ImmutableMap.Builder<PlanNodeId, List<PrestoSparkShuffleInput>> shuffleInputs = ImmutableMap.builder();
    ImmutableMap.Builder<PlanNodeId, List<java.util.Iterator<PrestoSparkSerializedPage>>> pageInputs = ImmutableMap.builder();
    ImmutableMap.Builder<PlanNodeId, List<?>> broadcastInputs = ImmutableMap.builder();
    for (RemoteSourceNode remoteSource : fragment.getRemoteSourceNodes()) {
        List<PrestoSparkShuffleInput> remoteSourceRowInputs = new ArrayList<>();
        List<java.util.Iterator<PrestoSparkSerializedPage>> remoteSourcePageInputs = new ArrayList<>();
        List<List<?>> broadcastInputsList = new ArrayList<>();
        for (PlanFragmentId sourceFragmentId : remoteSource.getSourceFragmentIds()) {
            Iterator<Tuple2<MutablePartitionId, PrestoSparkMutableRow>> shuffleInput = inputs.getShuffleInputs().get(sourceFragmentId.toString());
            Broadcast<?> broadcastInput = inputs.getBroadcastInputs().get(sourceFragmentId.toString());
            List<PrestoSparkSerializedPage> inMemoryInput = inputs.getInMemoryInputs().get(sourceFragmentId.toString());
            if (shuffleInput != null) {
                checkArgument(broadcastInput == null, "single remote source is not expected to accept different kind of inputs");
                checkArgument(inMemoryInput == null, "single remote source is not expected to accept different kind of inputs");
                remoteSourceRowInputs.add(new PrestoSparkShuffleInput(sourceFragmentId.getId(), shuffleInput));
                continue;
            }
            if (broadcastInput != null) {
                checkArgument(inMemoryInput == null, "single remote source is not expected to accept different kind of inputs");
                // TODO: Enable NullifyingIterator once migrated to one task per JVM model
                // NullifyingIterator removes element from the list upon return
                // This allows GC to gradually reclaim memory
                // remoteSourcePageInputs.add(getNullifyingIterator(broadcastInput.value()));
                broadcastInputsList.add((List<?>) broadcastInput.value());
                continue;
            }
            if (inMemoryInput != null) {
                // for inmemory inputs pages can be released incrementally to save memory
                remoteSourcePageInputs.add(getNullifyingIterator(inMemoryInput));
                continue;
            }
            throw new IllegalArgumentException("Input not found for sourceFragmentId: " + sourceFragmentId);
        }
        if (!remoteSourceRowInputs.isEmpty()) {
            shuffleInputs.put(remoteSource.getId(), remoteSourceRowInputs);
        }
        if (!remoteSourcePageInputs.isEmpty()) {
            pageInputs.put(remoteSource.getId(), remoteSourcePageInputs);
        }
        if (!broadcastInputsList.isEmpty()) {
            broadcastInputs.put(remoteSource.getId(), broadcastInputsList);
        }
    }
    OutputBufferMemoryManager memoryManager = new OutputBufferMemoryManager(sinkMaxBufferSize.toBytes(), () -> queryContext.getTaskContextByTaskId(taskId).localSystemMemoryContext(), notificationExecutor);
    Optional<OutputPartitioning> preDeterminedPartition = Optional.empty();
    if (fragment.getPartitioningScheme().getPartitioning().getHandle().equals(FIXED_ARBITRARY_DISTRIBUTION)) {
        int partitionCount = getHashPartitionCount(session);
        preDeterminedPartition = Optional.of(new OutputPartitioning(new PreDeterminedPartitionFunction(partitionId % partitionCount, partitionCount), ImmutableList.of(), ImmutableList.of(), false, OptionalInt.empty()));
    }
    TempDataOperationContext tempDataOperationContext = new TempDataOperationContext(session.getSource(), session.getQueryId().getId(), session.getClientInfo(), Optional.of(session.getClientTags()), session.getIdentity());
    TempStorage tempStorage = tempStorageManager.getTempStorage(storageBasedBroadcastJoinStorage);
    Output<T> output = configureOutput(outputType, blockEncodingManager, memoryManager, getShuffleOutputTargetAverageRowSize(session), preDeterminedPartition, tempStorage, tempDataOperationContext, getStorageBasedBroadcastJoinWriteBufferSize(session));
    PrestoSparkOutputBuffer<?> outputBuffer = output.getOutputBuffer();
    LocalExecutionPlan localExecutionPlan = localExecutionPlanner.plan(taskContext, fragment.getRoot(), fragment.getPartitioningScheme(), fragment.getStageExecutionDescriptor(), fragment.getTableScanSchedulingOrder(), output.getOutputFactory(), new PrestoSparkRemoteSourceFactory(blockEncodingManager, shuffleInputs.build(), pageInputs.build(), broadcastInputs.build(), partitionId, shuffleStatsCollector, tempStorage, tempDataOperationContext, prestoSparkBroadcastTableCacheManager, stageId), taskDescriptor.getTableWriteInfo(), true);
    taskStateMachine.addStateChangeListener(state -> {
        if (state.isDone()) {
            outputBuffer.setNoMoreRows();
        }
    });
    PrestoSparkTaskExecution taskExecution = new PrestoSparkTaskExecution(taskStateMachine, taskContext, localExecutionPlan, taskExecutor, splitMonitor, notificationExecutor, memoryUpdateExecutor);
    taskExecution.start(taskSources);
    return new PrestoSparkTaskExecutor<>(taskContext, taskStateMachine, output.getOutputSupplier(), taskInfoCodec, taskInfoCollector, shuffleStatsCollector, executionExceptionFactory, output.getOutputBufferType(), outputBuffer, tempStorage, tempDataOperationContext);
}

Example 2 with SpillSpaceTracker

use of com.facebook.presto.spiller.SpillSpaceTracker in project presto by prestodb.

the class GroupByHashYieldAssertion method finishOperatorWithYieldingGroupByHash.

/**
 * @param operatorFactory creates an Operator that should directly or indirectly contain GroupByHash
 * @param getHashCapacity returns the hash table capacity for the input operator
 * @param additionalMemoryInBytes the memory used in addition to the GroupByHash in the operator (e.g., aggregator)
 */
public static GroupByHashYieldResult finishOperatorWithYieldingGroupByHash(List<Page> input, Type hashKeyType, OperatorFactory operatorFactory, Function<Operator, Integer> getHashCapacity, long additionalMemoryInBytes) {
    assertLessThan(additionalMemoryInBytes, 1L << 21, "additionalMemoryInBytes should be a relatively small number");
    List<Page> result = new LinkedList<>();
    // mock an adjustable memory pool
    QueryId queryId1 = new QueryId("test_query1");
    QueryId queryId2 = new QueryId("test_query2");
    MemoryPool memoryPool = new MemoryPool(new MemoryPoolId("test"), new DataSize(1, GIGABYTE));
    QueryContext queryContext = new QueryContext(queryId2, new DataSize(512, MEGABYTE), new DataSize(1024, MEGABYTE), new DataSize(512, MEGABYTE), new DataSize(1, GIGABYTE), memoryPool, new TestingGcMonitor(), EXECUTOR, SCHEDULED_EXECUTOR, new DataSize(512, MEGABYTE), new SpillSpaceTracker(new DataSize(512, MEGABYTE)), listJsonCodec(TaskMemoryReservationSummary.class));
    DriverContext driverContext = createTaskContext(queryContext, EXECUTOR, TEST_SESSION).addPipelineContext(0, true, true, false).addDriverContext();
    Operator operator = operatorFactory.createOperator(driverContext);
    // run operator
    int yieldCount = 0;
    long expectedReservedExtraBytes = 0;
    for (Page page : input) {
        // unblocked
        assertTrue(operator.needsInput());
        // saturate the pool with a tiny memory left
        long reservedMemoryInBytes = memoryPool.getFreeBytes() - additionalMemoryInBytes;
        memoryPool.reserve(queryId1, "test", reservedMemoryInBytes);
        long oldMemoryUsage = operator.getOperatorContext().getDriverContext().getMemoryUsage();
        int oldCapacity = getHashCapacity.apply(operator);
        // add a page and verify different behaviors
        operator.addInput(page);
        // get output to consume the input
        Page output = operator.getOutput();
        if (output != null) {
            result.add(output);
        }
        long newMemoryUsage = operator.getOperatorContext().getDriverContext().getMemoryUsage();
        // between rehash and memory used by aggregator
        if (newMemoryUsage < new DataSize(4, MEGABYTE).toBytes()) {
            // free the pool for the next iteration
            memoryPool.free(queryId1, "test", reservedMemoryInBytes);
            // this required in case input is blocked
            operator.getOutput();
            continue;
        }
        long actualIncreasedMemory = newMemoryUsage - oldMemoryUsage;
        if (operator.needsInput()) {
            // We have successfully added a page
            // Assert we are not blocked
            assertTrue(operator.getOperatorContext().isWaitingForMemory().isDone());
            // assert the hash capacity is not changed; otherwise, we should have yielded
            assertTrue(oldCapacity == getHashCapacity.apply(operator));
            // We are not going to rehash; therefore, assert the memory increase only comes from the aggregator
            assertLessThan(actualIncreasedMemory, additionalMemoryInBytes);
            // free the pool for the next iteration
            memoryPool.free(queryId1, "test", reservedMemoryInBytes);
        } else {
            // We failed to finish the page processing i.e. we yielded
            yieldCount++;
            // Assert we are blocked
            assertFalse(operator.getOperatorContext().isWaitingForMemory().isDone());
            // Hash table capacity should not change
            assertEquals(oldCapacity, (long) getHashCapacity.apply(operator));
            // Increased memory is no smaller than the hash table size and no greater than the hash table size + the memory used by aggregator
            if (hashKeyType == BIGINT) {
                // groupIds and values double by hashCapacity; while valuesByGroupId double by maxFill = hashCapacity / 0.75
                expectedReservedExtraBytes = oldCapacity * (long) (Long.BYTES * 1.75 + Integer.BYTES) + page.getRetainedSizeInBytes();
            } else {
                // groupAddressByHash, groupIdsByHash, and rawHashByHashPosition double by hashCapacity; while groupAddressByGroupId double by maxFill = hashCapacity / 0.75
                expectedReservedExtraBytes = oldCapacity * (long) (Long.BYTES * 1.75 + Integer.BYTES + Byte.BYTES) + page.getRetainedSizeInBytes();
            }
            assertBetweenInclusive(actualIncreasedMemory, expectedReservedExtraBytes, expectedReservedExtraBytes + additionalMemoryInBytes);
            // Output should be blocked as well
            assertNull(operator.getOutput());
            // Free the pool to unblock
            memoryPool.free(queryId1, "test", reservedMemoryInBytes);
            // Trigger a process through getOutput() or needsInput()
            output = operator.getOutput();
            if (output != null) {
                result.add(output);
            }
            assertTrue(operator.needsInput());
            // Hash table capacity has increased
            assertGreaterThan(getHashCapacity.apply(operator), oldCapacity);
            // Assert the estimated reserved memory before rehash is very close to the one after rehash
            long rehashedMemoryUsage = operator.getOperatorContext().getDriverContext().getMemoryUsage();
            assertBetweenInclusive(rehashedMemoryUsage * 1.0 / newMemoryUsage, 0.99, 1.01);
            // unblocked
            assertTrue(operator.needsInput());
        }
    }
    result.addAll(finishOperator(operator));
    return new GroupByHashYieldResult(yieldCount, expectedReservedExtraBytes, result);
}

Example 3 with SpillSpaceTracker

use of com.facebook.presto.spiller.SpillSpaceTracker in project presto by prestodb.

the class TestQueryContext method testChecksTotalMemoryOnUserMemoryAllocation.

@Test(expectedExceptions = ExceededMemoryLimitException.class, expectedExceptionsMessageRegExp = ".*Query exceeded per-node total memory limit of 20B.*")
public void testChecksTotalMemoryOnUserMemoryAllocation() {
    try (LocalQueryRunner localQueryRunner = new LocalQueryRunner(TEST_SESSION)) {
        QueryContext queryContext = new QueryContext(new QueryId("query"), // user memory limit
        new DataSize(10, BYTE), // total memory limit
        new DataSize(20, BYTE), new DataSize(10, BYTE), new DataSize(1, GIGABYTE), new MemoryPool(GENERAL_POOL, new DataSize(10, BYTE)), new TestingGcMonitor(), localQueryRunner.getExecutor(), localQueryRunner.getScheduler(), new DataSize(0, BYTE), new SpillSpaceTracker(new DataSize(0, BYTE)), listJsonCodec(TaskMemoryReservationSummary.class));
        queryContext.getQueryMemoryContext().initializeLocalMemoryContexts("test");
        LocalMemoryContext systemMemoryContext = queryContext.getQueryMemoryContext().localSystemMemoryContext();
        LocalMemoryContext userMemoryContext = queryContext.getQueryMemoryContext().localUserMemoryContext();
        systemMemoryContext.setBytes(15);
        userMemoryContext.setBytes(6);
    }
}

Example 4 with SpillSpaceTracker

use of com.facebook.presto.spiller.SpillSpaceTracker in project presto by prestodb.

the class TestMemoryPools method setUp.

private void setUp(Supplier<List<Driver>> driversSupplier) {
    checkState(localQueryRunner == null, "Already set up");
    Session session = testSessionBuilder().setCatalog("tpch").setSchema("tiny").setSystemProperty("task_default_concurrency", "1").build();
    localQueryRunner = queryRunnerWithInitialTransaction(session);
    // add tpch
    localQueryRunner.createCatalog("tpch", new TpchConnectorFactory(1), ImmutableMap.of());
    userPool = new MemoryPool(new MemoryPoolId("test"), TEN_MEGABYTES);
    fakeQueryId = new QueryId("fake");
    SpillSpaceTracker spillSpaceTracker = new SpillSpaceTracker(new DataSize(1, GIGABYTE));
    QueryContext queryContext = new QueryContext(new QueryId("query"), TEN_MEGABYTES, new DataSize(20, MEGABYTE), TEN_MEGABYTES, new DataSize(1, GIGABYTE), userPool, new TestingGcMonitor(), localQueryRunner.getExecutor(), localQueryRunner.getScheduler(), TEN_MEGABYTES, spillSpaceTracker, listJsonCodec(TaskMemoryReservationSummary.class));
    taskContext = createTaskContext(queryContext, localQueryRunner.getExecutor(), session);
    drivers = driversSupplier.get();
}

Example 5 with SpillSpaceTracker

use of com.facebook.presto.spiller.SpillSpaceTracker in project presto by prestodb.

the class MemoryLocalQueryRunner method execute.

public List<Page> execute(@Language("SQL") String query) {
    MemoryPool memoryPool = new MemoryPool(new MemoryPoolId("test"), new DataSize(2, GIGABYTE));
    SpillSpaceTracker spillSpaceTracker = new SpillSpaceTracker(new DataSize(1, GIGABYTE));
    QueryContext queryContext = new QueryContext(new QueryId("test"), new DataSize(1, GIGABYTE), new DataSize(2, GIGABYTE), new DataSize(1, GIGABYTE), new DataSize(2, GIGABYTE), memoryPool, new TestingGcMonitor(), localQueryRunner.getExecutor(), localQueryRunner.getScheduler(), new DataSize(4, GIGABYTE), spillSpaceTracker, listJsonCodec(TaskMemoryReservationSummary.class));
    TaskContext taskContext = queryContext.addTaskContext(new TaskStateMachine(new TaskId("query", 0, 0, 0), localQueryRunner.getExecutor()), localQueryRunner.getDefaultSession(), Optional.empty(), false, false, false, false, false);
    // Use NullOutputFactory to avoid coping out results to avoid affecting benchmark results
    ImmutableList.Builder<Page> output = ImmutableList.builder();
    List<Driver> drivers = localQueryRunner.createDrivers(query, new PageConsumerOperator.PageConsumerOutputFactory(types -> output::add), taskContext);
    boolean done = false;
    while (!done) {
        boolean processed = false;
        for (Driver driver : drivers) {
            if (!driver.isFinished()) {
                driver.process();
                processed = true;
            }
        }
        done = !processed;
    }
    return output.build();
}