ClickHouse/src/Interpreters/Aggregator.cpp

2500 lines
87 KiB
C++
Raw Normal View History

#include <iomanip>
#include <thread>
#include <future>
#include <Poco/Version.h>
#include <Poco/Util/Application.h>
#include <Common/Stopwatch.h>
#include <Common/setThreadName.h>
2020-05-23 20:57:48 +00:00
#include <Common/formatReadable.h>
#include <DataTypes/DataTypeAggregateFunction.h>
#include <DataTypes/DataTypeNullable.h>
2019-01-21 10:39:24 +00:00
#include <DataTypes/DataTypeLowCardinality.h>
#include <Columns/ColumnsNumber.h>
#include <Columns/ColumnArray.h>
#include <Columns/ColumnTuple.h>
2019-01-21 10:39:24 +00:00
#include <Columns/ColumnLowCardinality.h>
#include <DataStreams/IBlockInputStream.h>
#include <DataStreams/NativeBlockOutputStream.h>
#include <DataStreams/NullBlockInputStream.h>
2018-02-21 06:25:21 +00:00
#include <DataStreams/materializeBlock.h>
#include <IO/WriteBufferFromFile.h>
2018-12-28 18:15:26 +00:00
#include <Compression/CompressedWriteBuffer.h>
#include <Interpreters/Aggregator.h>
#include <Common/ClickHouseRevision.h>
2017-04-08 01:32:05 +00:00
#include <Common/MemoryTracker.h>
2018-05-29 18:14:31 +00:00
#include <Common/CurrentThread.h>
2017-07-13 20:58:19 +00:00
#include <Common/typeid_cast.h>
#include <Common/assert_cast.h>
#include <common/demangle.h>
#include <AggregateFunctions/AggregateFunctionArray.h>
#include <AggregateFunctions/AggregateFunctionState.h>
#include <Disks/StoragePolicy.h>
2019-01-21 10:39:24 +00:00
2011-09-19 01:42:16 +00:00
namespace ProfileEvents
{
extern const Event ExternalAggregationWritePart;
extern const Event ExternalAggregationCompressedBytes;
extern const Event ExternalAggregationUncompressedBytes;
}
namespace CurrentMetrics
{
extern const Metric QueryThread;
}
2011-09-19 01:42:16 +00:00
namespace DB
{
namespace ErrorCodes
{
2020-02-25 18:02:41 +00:00
extern const int UNKNOWN_AGGREGATED_DATA_VARIANT;
extern const int NOT_ENOUGH_SPACE;
2018-03-09 23:23:15 +00:00
extern const int TOO_MANY_ROWS;
extern const int EMPTY_DATA_PASSED;
extern const int CANNOT_MERGE_DIFFERENT_AGGREGATED_DATA_VARIANTS;
2018-09-01 03:17:43 +00:00
extern const int LOGICAL_ERROR;
}
2011-09-26 07:25:22 +00:00
AggregatedDataVariants::~AggregatedDataVariants()
{
if (aggregator && !aggregator->all_aggregates_has_trivial_destructor)
{
try
{
aggregator->destroyAllAggregateStates(*this);
}
catch (...)
{
tryLogCurrentException(__PRETTY_FUNCTION__);
}
}
}
void AggregatedDataVariants::convertToTwoLevel()
{
if (aggregator)
2020-05-23 22:24:01 +00:00
LOG_TRACE(aggregator->log, "Converting aggregation data to two-level.");
switch (type)
{
#define M(NAME) \
case Type::NAME: \
2020-03-18 02:02:24 +00:00
NAME ## _two_level = std::make_unique<decltype(NAME ## _two_level)::element_type>(*(NAME)); \
(NAME).reset(); \
type = Type::NAME ## _two_level; \
break;
APPLY_FOR_VARIANTS_CONVERTIBLE_TO_TWO_LEVEL(M)
#undef M
default:
throw Exception("Wrong data variant passed.", ErrorCodes::LOGICAL_ERROR);
}
}
Block Aggregator::getHeader(bool final) const
{
Block res;
if (params.src_header)
{
for (size_t i = 0; i < params.keys_size; ++i)
res.insert(params.src_header.safeGetByPosition(params.keys[i]).cloneEmpty());
for (size_t i = 0; i < params.aggregates_size; ++i)
{
size_t arguments_size = params.aggregates[i].arguments.size();
DataTypes argument_types(arguments_size);
for (size_t j = 0; j < arguments_size; ++j)
argument_types[j] = params.src_header.safeGetByPosition(params.aggregates[i].arguments[j]).type;
DataTypePtr type;
if (final)
type = params.aggregates[i].function->getReturnType();
else
type = std::make_shared<DataTypeAggregateFunction>(params.aggregates[i].function, argument_types, params.aggregates[i].parameters);
2018-02-26 03:37:08 +00:00
res.insert({ type, params.aggregates[i].column_name });
}
}
else if (params.intermediate_header)
{
res = params.intermediate_header.cloneEmpty();
if (final)
{
for (const auto & aggregate : params.aggregates)
{
auto & elem = res.getByName(aggregate.column_name);
elem.type = aggregate.function->getReturnType();
elem.column = elem.type->createColumn();
}
}
}
2018-02-21 06:25:21 +00:00
return materializeBlock(res);
}
Aggregator::Aggregator(const Params & params_)
: params(params_),
isCancelled([]() { return false; })
{
/// Use query-level memory tracker
2020-04-22 05:39:31 +00:00
if (auto * memory_tracker_child = CurrentThread::getMemoryTracker())
if (auto * memory_tracker = memory_tracker_child->getParent())
memory_usage_before_aggregation = memory_tracker->get();
aggregate_functions.resize(params.aggregates_size);
for (size_t i = 0; i < params.aggregates_size; ++i)
aggregate_functions[i] = params.aggregates[i].function.get();
/// Initialize sizes of aggregation states and its offsets.
offsets_of_aggregate_states.resize(params.aggregates_size);
total_size_of_aggregate_states = 0;
all_aggregates_has_trivial_destructor = true;
2020-01-11 09:50:41 +00:00
// aggregate_states will be aligned as below:
2018-08-05 08:45:15 +00:00
// |<-- state_1 -->|<-- pad_1 -->|<-- state_2 -->|<-- pad_2 -->| .....
//
// pad_N will be used to match alignment requirement for each next state.
// The address of state_1 is aligned based on maximum alignment requirements in states
for (size_t i = 0; i < params.aggregates_size; ++i)
{
offsets_of_aggregate_states[i] = total_size_of_aggregate_states;
2018-08-05 08:45:15 +00:00
total_size_of_aggregate_states += params.aggregates[i].function->sizeOfData();
2020-01-11 09:50:41 +00:00
// aggregate states are aligned based on maximum requirement
2018-09-01 03:17:43 +00:00
align_aggregate_states = std::max(align_aggregate_states, params.aggregates[i].function->alignOfData());
2018-08-05 08:45:15 +00:00
2018-09-01 03:17:43 +00:00
// If not the last aggregate_state, we need pad it so that next aggregate_state will be aligned.
2018-08-05 08:45:15 +00:00
if (i + 1 < params.aggregates_size)
{
2018-09-01 03:17:43 +00:00
size_t alignment_of_next_state = params.aggregates[i + 1].function->alignOfData();
if ((alignment_of_next_state & (alignment_of_next_state - 1)) != 0)
throw Exception("Logical error: alignOfData is not 2^N", ErrorCodes::LOGICAL_ERROR);
/// Extend total_size to next alignment requirement
/// Add padding by rounding up 'total_size_of_aggregate_states' to be a multiplier of alignment_of_next_state.
total_size_of_aggregate_states = (total_size_of_aggregate_states + alignment_of_next_state - 1) / alignment_of_next_state * alignment_of_next_state;
2018-08-05 08:45:15 +00:00
}
if (!params.aggregates[i].function->hasTrivialDestructor())
all_aggregates_has_trivial_destructor = false;
}
2018-08-27 17:42:13 +00:00
method_chosen = chooseAggregationMethod();
2019-01-21 10:39:24 +00:00
HashMethodContext::Settings cache_settings;
cache_settings.max_threads = params.max_threads;
2018-09-14 09:14:37 +00:00
aggregation_state_cache = AggregatedDataVariants::createCache(method_chosen, cache_settings);
}
AggregatedDataVariants::Type Aggregator::chooseAggregationMethod()
2012-05-30 01:38:02 +00:00
{
/// If no keys. All aggregating to single row.
if (params.keys_size == 0)
return AggregatedDataVariants::Type::without_key;
/// Check if at least one of the specified keys is nullable.
DataTypes types_removed_nullable;
types_removed_nullable.reserve(params.keys.size());
bool has_nullable_key = false;
bool has_low_cardinality = false;
for (const auto & pos : params.keys)
{
DataTypePtr type = (params.src_header ? params.src_header : params.intermediate_header).safeGetByPosition(pos).type;
if (type->lowCardinality())
{
has_low_cardinality = true;
type = removeLowCardinality(type);
}
if (type->isNullable())
{
has_nullable_key = true;
type = removeNullable(type);
}
types_removed_nullable.push_back(type);
}
/** Returns ordinary (not two-level) methods, because we start from them.
* Later, during aggregation process, data may be converted (partitioned) to two-level structure, if cardinality is high.
*/
size_t keys_bytes = 0;
size_t num_fixed_contiguous_keys = 0;
key_sizes.resize(params.keys_size);
for (size_t j = 0; j < params.keys_size; ++j)
{
if (types_removed_nullable[j]->isValueUnambiguouslyRepresentedInContiguousMemoryRegion())
{
if (types_removed_nullable[j]->isValueUnambiguouslyRepresentedInFixedSizeContiguousMemoryRegion())
{
++num_fixed_contiguous_keys;
key_sizes[j] = types_removed_nullable[j]->getSizeOfValueInMemory();
keys_bytes += key_sizes[j];
}
}
}
if (has_nullable_key)
{
if (params.keys_size == num_fixed_contiguous_keys && !has_low_cardinality)
{
/// Pack if possible all the keys along with information about which key values are nulls
/// into a fixed 16- or 32-byte blob.
if (std::tuple_size<KeysNullMap<UInt128>>::value + keys_bytes <= 16)
return AggregatedDataVariants::Type::nullable_keys128;
if (std::tuple_size<KeysNullMap<UInt256>>::value + keys_bytes <= 32)
return AggregatedDataVariants::Type::nullable_keys256;
}
if (has_low_cardinality && params.keys_size == 1)
{
if (types_removed_nullable[0]->isValueRepresentedByNumber())
{
size_t size_of_field = types_removed_nullable[0]->getSizeOfValueInMemory();
if (size_of_field == 1)
return AggregatedDataVariants::Type::low_cardinality_key8;
if (size_of_field == 2)
return AggregatedDataVariants::Type::low_cardinality_key16;
if (size_of_field == 4)
return AggregatedDataVariants::Type::low_cardinality_key32;
if (size_of_field == 8)
return AggregatedDataVariants::Type::low_cardinality_key64;
}
else if (isString(types_removed_nullable[0]))
return AggregatedDataVariants::Type::low_cardinality_key_string;
else if (isFixedString(types_removed_nullable[0]))
return AggregatedDataVariants::Type::low_cardinality_key_fixed_string;
}
/// Fallback case.
return AggregatedDataVariants::Type::serialized;
}
/// No key has been found to be nullable.
/// Single numeric key.
if (params.keys_size == 1 && types_removed_nullable[0]->isValueRepresentedByNumber())
{
size_t size_of_field = types_removed_nullable[0]->getSizeOfValueInMemory();
if (has_low_cardinality)
{
if (size_of_field == 1)
return AggregatedDataVariants::Type::low_cardinality_key8;
if (size_of_field == 2)
return AggregatedDataVariants::Type::low_cardinality_key16;
if (size_of_field == 4)
return AggregatedDataVariants::Type::low_cardinality_key32;
if (size_of_field == 8)
return AggregatedDataVariants::Type::low_cardinality_key64;
}
if (size_of_field == 1)
return AggregatedDataVariants::Type::key8;
if (size_of_field == 2)
return AggregatedDataVariants::Type::key16;
if (size_of_field == 4)
return AggregatedDataVariants::Type::key32;
if (size_of_field == 8)
return AggregatedDataVariants::Type::key64;
if (size_of_field == 16)
return AggregatedDataVariants::Type::keys128;
throw Exception("Logical error: numeric column has sizeOfField not in 1, 2, 4, 8, 16.", ErrorCodes::LOGICAL_ERROR);
}
/// If all keys fits in N bits, will use hash table with all keys packed (placed contiguously) to single N-bit key.
if (params.keys_size == num_fixed_contiguous_keys)
{
if (has_low_cardinality)
{
if (keys_bytes <= 16)
return AggregatedDataVariants::Type::low_cardinality_keys128;
if (keys_bytes <= 32)
return AggregatedDataVariants::Type::low_cardinality_keys256;
}
2020-03-23 13:44:21 +00:00
if (keys_bytes <= 2)
return AggregatedDataVariants::Type::keys16;
if (keys_bytes <= 4)
return AggregatedDataVariants::Type::keys32;
if (keys_bytes <= 8)
return AggregatedDataVariants::Type::keys64;
if (keys_bytes <= 16)
return AggregatedDataVariants::Type::keys128;
if (keys_bytes <= 32)
return AggregatedDataVariants::Type::keys256;
}
/// If single string key - will use hash table with references to it. Strings itself are stored separately in Arena.
if (params.keys_size == 1 && isString(types_removed_nullable[0]))
{
if (has_low_cardinality)
return AggregatedDataVariants::Type::low_cardinality_key_string;
else
return AggregatedDataVariants::Type::key_string;
}
if (params.keys_size == 1 && isFixedString(types_removed_nullable[0]))
{
if (has_low_cardinality)
return AggregatedDataVariants::Type::low_cardinality_key_fixed_string;
else
return AggregatedDataVariants::Type::key_fixed_string;
}
return AggregatedDataVariants::Type::serialized;
2012-05-30 01:38:02 +00:00
}
void Aggregator::createAggregateStates(AggregateDataPtr & aggregate_data) const
{
for (size_t j = 0; j < params.aggregates_size; ++j)
{
try
{
2017-04-02 17:37:49 +00:00
/** An exception may occur if there is a shortage of memory.
* In order that then everything is properly destroyed, we "roll back" some of the created states.
* The code is not very convenient.
*/
aggregate_functions[j]->create(aggregate_data + offsets_of_aggregate_states[j]);
}
catch (...)
{
for (size_t rollback_j = 0; rollback_j < j; ++rollback_j)
aggregate_functions[rollback_j]->destroy(aggregate_data + offsets_of_aggregate_states[rollback_j]);
throw;
}
}
}
2017-04-02 17:37:49 +00:00
/** It's interesting - if you remove `noinline`, then gcc for some reason will inline this function, and the performance decreases (~ 10%).
* (Probably because after the inline of this function, more internal functions no longer be inlined.)
* Inline does not make sense, since the inner loop is entirely inside this function.
*/
template <typename Method>
void NO_INLINE Aggregator::executeImpl(
Method & method,
Arena * aggregates_pool,
size_t rows,
ColumnRawPtrs & key_columns,
AggregateFunctionInstruction * aggregate_instructions,
bool no_more_keys,
AggregateDataPtr overflow_row) const
{
2019-01-21 10:39:24 +00:00
typename Method::State state(key_columns, key_sizes, aggregation_state_cache);
if (!no_more_keys)
2019-08-10 22:36:55 +00:00
//executeImplCase<false>(method, state, aggregates_pool, rows, aggregate_instructions, overflow_row);
executeImplBatch(method, state, aggregates_pool, rows, aggregate_instructions);
else
2019-08-10 22:36:55 +00:00
executeImplCase<true>(method, state, aggregates_pool, rows, aggregate_instructions, overflow_row);
}
template <bool no_more_keys, typename Method>
void NO_INLINE Aggregator::executeImplCase(
Method & method,
typename Method::State & state,
Arena * aggregates_pool,
size_t rows,
AggregateFunctionInstruction * aggregate_instructions,
AggregateDataPtr overflow_row) const
{
2017-04-02 17:37:49 +00:00
/// NOTE When editing this code, also pay attention to SpecializedAggregator.h.
2017-04-02 17:37:49 +00:00
/// For all rows.
for (size_t i = 0; i < rows; ++i)
{
2019-01-21 10:39:53 +00:00
AggregateDataPtr aggregate_data = nullptr;
2019-01-21 10:39:24 +00:00
if constexpr (!no_more_keys) /// Insert.
2019-01-21 10:39:53 +00:00
{
auto emplace_result = state.emplaceKey(method.data, i, *aggregates_pool);
/// If a new key is inserted, initialize the states of the aggregate functions, and possibly something related to the key.
if (emplace_result.isInserted())
{
/// exception-safety - if you can not allocate memory or create states, then destructors will not be called.
emplace_result.setMapped(nullptr);
aggregate_data = aggregates_pool->alignedAlloc(total_size_of_aggregate_states, align_aggregate_states);
createAggregateStates(aggregate_data);
emplace_result.setMapped(aggregate_data);
}
else
aggregate_data = emplace_result.getMapped();
}
else
{
2017-04-02 17:37:49 +00:00
/// Add only if the key already exists.
2019-01-21 10:39:53 +00:00
auto find_result = state.findKey(method.data, i, *aggregates_pool);
if (find_result.isFound())
aggregate_data = find_result.getMapped();
}
/// aggregate_date == nullptr means that the new key did not fit in the hash table because of no_more_keys.
2017-04-02 17:37:49 +00:00
/// If the key does not fit, and the data does not need to be aggregated in a separate row, then there's nothing to do.
if (!aggregate_data && !overflow_row)
continue;
2019-01-21 10:39:53 +00:00
AggregateDataPtr value = aggregate_data ? aggregate_data : overflow_row;
2017-04-02 17:37:49 +00:00
/// Add values to the aggregate functions.
for (AggregateFunctionInstruction * inst = aggregate_instructions; inst->that; ++inst)
2019-11-11 08:36:19 +00:00
(*inst->func)(inst->that, value + inst->state_offset, inst->arguments, i, aggregates_pool);
}
}
2019-08-10 22:36:55 +00:00
template <typename Method>
void NO_INLINE Aggregator::executeImplBatch(
Method & method,
typename Method::State & state,
Arena * aggregates_pool,
size_t rows,
AggregateFunctionInstruction * aggregate_instructions) const
{
PODArray<AggregateDataPtr> places(rows);
/// For all rows.
for (size_t i = 0; i < rows; ++i)
{
AggregateDataPtr aggregate_data = nullptr;
auto emplace_result = state.emplaceKey(method.data, i, *aggregates_pool);
/// If a new key is inserted, initialize the states of the aggregate functions, and possibly something related to the key.
if (emplace_result.isInserted())
{
/// exception-safety - if you can not allocate memory or create states, then destructors will not be called.
emplace_result.setMapped(nullptr);
aggregate_data = aggregates_pool->alignedAlloc(total_size_of_aggregate_states, align_aggregate_states);
createAggregateStates(aggregate_data);
emplace_result.setMapped(aggregate_data);
}
else
aggregate_data = emplace_result.getMapped();
places[i] = aggregate_data;
assert(places[i] != nullptr);
2019-08-10 22:36:55 +00:00
}
/// Add values to the aggregate functions.
for (AggregateFunctionInstruction * inst = aggregate_instructions; inst->that; ++inst)
{
2019-11-11 08:36:19 +00:00
if (inst->offsets)
inst->batch_that->addBatchArray(rows, places.data(), inst->state_offset, inst->batch_arguments, inst->offsets, aggregates_pool);
else
inst->batch_that->addBatch(rows, places.data(), inst->state_offset, inst->batch_arguments, aggregates_pool);
}
2019-08-10 22:36:55 +00:00
}
void NO_INLINE Aggregator::executeWithoutKeyImpl(
AggregatedDataWithoutKey & res,
size_t rows,
AggregateFunctionInstruction * aggregate_instructions,
2020-03-18 00:57:00 +00:00
Arena * arena)
{
/// Adding values
for (AggregateFunctionInstruction * inst = aggregate_instructions; inst->that; ++inst)
{
if (inst->offsets)
2019-11-11 08:36:19 +00:00
inst->batch_that->addBatchSinglePlace(
inst->offsets[static_cast<ssize_t>(rows - 1)], res + inst->state_offset, inst->batch_arguments, arena);
else
2019-11-11 08:36:19 +00:00
inst->batch_that->addBatchSinglePlace(rows, res + inst->state_offset, inst->batch_arguments, arena);
}
}
2020-04-18 09:51:21 +00:00
void NO_INLINE Aggregator::executeOnIntervalWithoutKeyImpl(
AggregatedDataWithoutKey & res,
size_t row_begin,
size_t row_end,
AggregateFunctionInstruction * aggregate_instructions,
Arena * arena)
2019-08-31 08:58:16 +00:00
{
2020-04-18 09:51:21 +00:00
/// Adding values
for (AggregateFunctionInstruction * inst = aggregate_instructions; inst->that; ++inst)
{
2020-04-18 09:51:21 +00:00
if (inst->offsets)
2020-05-13 13:49:10 +00:00
inst->batch_that->addBatchSinglePlaceFromInterval(inst->offsets[row_begin], inst->offsets[row_end - 1], res + inst->state_offset, inst->batch_arguments, arena);
2020-04-18 09:51:21 +00:00
else
inst->batch_that->addBatchSinglePlaceFromInterval(row_begin, row_end, res + inst->state_offset, inst->batch_arguments, arena);
}
2020-04-18 09:51:21 +00:00
}
2020-05-07 20:13:51 +00:00
void Aggregator::prepareAggregateInstructions(Columns columns, AggregateColumns & aggregate_columns, Columns & materialized_columns,
2020-05-15 19:27:18 +00:00
AggregateFunctionInstructions & aggregate_functions_instructions, NestedColumnsHolder & nested_columns_holder)
2020-04-18 09:51:21 +00:00
{
for (size_t i = 0; i < params.aggregates_size; ++i)
aggregate_columns[i].resize(params.aggregates[i].arguments.size());
2020-05-07 14:54:15 +00:00
aggregate_functions_instructions.resize(params.aggregates_size + 1);
aggregate_functions_instructions[params.aggregates_size].that = nullptr;
for (size_t i = 0; i < params.aggregates_size; ++i)
{
for (size_t j = 0; j < aggregate_columns[i].size(); ++j)
{
2019-08-31 08:58:16 +00:00
materialized_columns.push_back(columns.at(params.aggregates[i].arguments[j])->convertToFullColumnIfConst());
aggregate_columns[i][j] = materialized_columns.back().get();
auto column_no_lc = recursiveRemoveLowCardinality(aggregate_columns[i][j]->getPtr());
if (column_no_lc.get() != aggregate_columns[i][j])
{
materialized_columns.emplace_back(std::move(column_no_lc));
aggregate_columns[i][j] = materialized_columns.back().get();
}
}
aggregate_functions_instructions[i].arguments = aggregate_columns[i].data();
aggregate_functions_instructions[i].state_offset = offsets_of_aggregate_states[i];
2020-04-22 05:39:31 +00:00
auto * that = aggregate_functions[i];
/// Unnest consecutive trailing -State combinators
2020-04-22 05:39:31 +00:00
while (const auto * func = typeid_cast<const AggregateFunctionState *>(that))
that = func->getNestedFunction().get();
aggregate_functions_instructions[i].that = that;
2019-11-11 08:36:19 +00:00
aggregate_functions_instructions[i].func = that->getAddressOfAddFunction();
2020-04-22 05:39:31 +00:00
if (const auto * func = typeid_cast<const AggregateFunctionArray *>(that))
{
/// Unnest consecutive -State combinators before -Array
that = func->getNestedFunction().get();
2020-04-22 05:39:31 +00:00
while (const auto * nested_func = typeid_cast<const AggregateFunctionState *>(that))
that = nested_func->getNestedFunction().get();
auto [nested_columns, offsets] = checkAndGetNestedArrayOffset(aggregate_columns[i].data(), that->getArgumentTypes().size());
nested_columns_holder.push_back(std::move(nested_columns));
aggregate_functions_instructions[i].batch_arguments = nested_columns_holder.back().data();
aggregate_functions_instructions[i].offsets = offsets;
}
else
aggregate_functions_instructions[i].batch_arguments = aggregate_columns[i].data();
aggregate_functions_instructions[i].batch_that = that;
}
2020-04-18 09:51:21 +00:00
}
2020-05-08 13:28:18 +00:00
bool Aggregator::executeOnBlock(const Block & block, AggregatedDataVariants & result,
ColumnRawPtrs & key_columns, AggregateColumns & aggregate_columns, bool & no_more_keys)
{
UInt64 num_rows = block.rows();
return executeOnBlock(block.getColumns(), num_rows, result, key_columns, aggregate_columns, no_more_keys);
}
2019-08-31 08:58:16 +00:00
bool Aggregator::executeOnBlock(Columns columns, UInt64 num_rows, AggregatedDataVariants & result,
ColumnRawPtrs & key_columns, AggregateColumns & aggregate_columns, bool & no_more_keys)
2014-05-10 05:16:23 +00:00
{
if (isCancelled())
return true;
2017-04-02 17:37:49 +00:00
/// `result` will destroy the states of aggregate functions in the destructor
result.aggregator = this;
/// How to perform the aggregation?
if (result.empty())
{
2018-09-14 09:14:37 +00:00
result.init(method_chosen);
result.keys_size = params.keys_size;
result.key_sizes = key_sizes;
LOG_TRACE(log, "Aggregation method: {}", result.getMethodName());
}
if (isCancelled())
return true;
2017-04-02 17:37:49 +00:00
/** Constant columns are not supported directly during aggregation.
* To make them work anyway, we materialize them.
*/
Columns materialized_columns;
2020-05-07 20:13:51 +00:00
/// Remember the columns we will work with
for (size_t i = 0; i < params.keys_size; ++i)
{
materialized_columns.push_back(columns.at(params.keys[i])->convertToFullColumnIfConst());
key_columns[i] = materialized_columns.back().get();
if (!result.isLowCardinality())
{
auto column_no_lc = recursiveRemoveLowCardinality(key_columns[i]->getPtr());
if (column_no_lc.get() != key_columns[i])
{
materialized_columns.emplace_back(std::move(column_no_lc));
key_columns[i] = materialized_columns.back().get();
}
}
}
2020-05-15 19:27:18 +00:00
NestedColumnsHolder nested_columns_holder;
2020-05-07 20:13:51 +00:00
AggregateFunctionInstructions aggregate_functions_instructions;
2020-05-15 19:27:18 +00:00
prepareAggregateInstructions(columns, aggregate_columns, materialized_columns, aggregate_functions_instructions, nested_columns_holder);
if (isCancelled())
return true;
if ((params.overflow_row || result.type == AggregatedDataVariants::Type::without_key) && !result.without_key)
{
2018-09-01 03:17:43 +00:00
AggregateDataPtr place = result.aggregates_pool->alignedAlloc(total_size_of_aggregate_states, align_aggregate_states);
createAggregateStates(place);
result.without_key = place;
}
2017-04-02 17:37:49 +00:00
/// We select one of the aggregation methods and call it.
2017-04-02 17:37:49 +00:00
/// For the case when there are no keys (all aggregate into one row).
if (result.type == AggregatedDataVariants::Type::without_key)
{
2019-08-31 08:58:16 +00:00
executeWithoutKeyImpl(result.without_key, num_rows, aggregate_functions_instructions.data(), result.aggregates_pool);
}
else
{
2017-04-02 17:37:49 +00:00
/// This is where data is written that does not fit in `max_rows_to_group_by` with `group_by_overflow_mode = any`.
AggregateDataPtr overflow_row_ptr = params.overflow_row ? result.without_key : nullptr;
#define M(NAME, IS_TWO_LEVEL) \
else if (result.type == AggregatedDataVariants::Type::NAME) \
2019-08-31 08:58:16 +00:00
executeImpl(*result.NAME, result.aggregates_pool, num_rows, key_columns, aggregate_functions_instructions.data(), \
2019-08-10 22:36:55 +00:00
no_more_keys, overflow_row_ptr);
2020-03-08 23:48:08 +00:00
if (false) {} // NOLINT
2019-08-23 21:10:26 +00:00
APPLY_FOR_AGGREGATED_VARIANTS(M)
#undef M
}
size_t result_size = result.sizeWithoutOverflowRow();
Int64 current_memory_usage = 0;
2020-04-22 05:39:31 +00:00
if (auto * memory_tracker_child = CurrentThread::getMemoryTracker())
if (auto * memory_tracker = memory_tracker_child->getParent())
current_memory_usage = memory_tracker->get();
2020-06-07 21:05:36 +00:00
/// Here all the results in the sum are taken into account, from different threads.
auto result_size_bytes = current_memory_usage - memory_usage_before_aggregation;
bool worth_convert_to_two_level
= (params.group_by_two_level_threshold && result_size >= params.group_by_two_level_threshold)
|| (params.group_by_two_level_threshold_bytes && result_size_bytes >= static_cast<Int64>(params.group_by_two_level_threshold_bytes));
2017-04-02 17:37:49 +00:00
/** Converting to a two-level data structure.
* It allows you to make, in the subsequent, an effective merge - either economical from memory or parallel.
*/
if (result.isConvertibleToTwoLevel() && worth_convert_to_two_level)
result.convertToTwoLevel();
2017-04-02 17:37:49 +00:00
/// Checking the constraints.
if (!checkLimits(result_size, no_more_keys))
return false;
2017-04-02 17:37:49 +00:00
/** Flush data to disk if too much RAM is consumed.
* Data can only be flushed to disk if a two-level aggregation structure is used.
*/
if (params.max_bytes_before_external_group_by
&& result.isTwoLevel()
&& current_memory_usage > static_cast<Int64>(params.max_bytes_before_external_group_by)
&& worth_convert_to_two_level)
{
size_t size = current_memory_usage + params.min_free_disk_space;
const std::string tmp_path = params.tmp_volume->getNextDisk()->getPath();
// enoughSpaceInDirectory() is not enough to make it right, since
// another process (or another thread of aggregator) can consume all
// space.
//
// But true reservation (IVolume::reserve()) cannot be used here since
// current_memory_usage does not takes compression into account and
// will reserve way more that actually will be used.
//
// Hence let's do a simple check.
if (!enoughSpaceInDirectory(tmp_path, size))
throw Exception("Not enough space for external aggregation in " + tmp_path, ErrorCodes::NOT_ENOUGH_SPACE);
2019-08-27 18:59:21 +00:00
writeToTemporaryFile(result, tmp_path);
}
return true;
}
void Aggregator::writeToTemporaryFile(AggregatedDataVariants & data_variants, const String & tmp_path)
{
Stopwatch watch;
size_t rows = data_variants.size();
auto file = createTemporaryFile(tmp_path);
const std::string & path = file->path();
WriteBufferFromFile file_buf(path);
CompressedWriteBuffer compressed_buf(file_buf);
NativeBlockOutputStream block_out(compressed_buf, ClickHouseRevision::get(), getHeader(false));
2020-05-23 22:24:01 +00:00
LOG_DEBUG(log, "Writing part of aggregation data into temporary file {}.", path);
ProfileEvents::increment(ProfileEvents::ExternalAggregationWritePart);
/// Flush only two-level data and possibly overflow data.
#define M(NAME) \
else if (data_variants.type == AggregatedDataVariants::Type::NAME) \
2017-12-01 21:13:25 +00:00
writeToTemporaryFileImpl(data_variants, *data_variants.NAME, block_out);
2020-03-08 23:48:08 +00:00
if (false) {} // NOLINT
APPLY_FOR_VARIANTS_TWO_LEVEL(M)
#undef M
else
throw Exception("Unknown aggregated data variant.", ErrorCodes::UNKNOWN_AGGREGATED_DATA_VARIANT);
2017-04-02 17:37:49 +00:00
/// NOTE Instead of freeing up memory and creating new hash tables and arenas, you can re-use the old ones.
data_variants.init(data_variants.type);
data_variants.aggregates_pools = Arenas(1, std::make_shared<Arena>());
data_variants.aggregates_pool = data_variants.aggregates_pools.back().get();
data_variants.without_key = nullptr;
block_out.flush();
compressed_buf.next();
file_buf.next();
double elapsed_seconds = watch.elapsedSeconds();
double compressed_bytes = file_buf.count();
double uncompressed_bytes = compressed_buf.count();
{
2019-01-02 06:44:36 +00:00
std::lock_guard lock(temporary_files.mutex);
temporary_files.files.emplace_back(std::move(file));
temporary_files.sum_size_uncompressed += uncompressed_bytes;
temporary_files.sum_size_compressed += compressed_bytes;
}
ProfileEvents::increment(ProfileEvents::ExternalAggregationCompressedBytes, compressed_bytes);
ProfileEvents::increment(ProfileEvents::ExternalAggregationUncompressedBytes, uncompressed_bytes);
2020-05-23 22:24:01 +00:00
LOG_TRACE(log,
2020-05-23 21:16:05 +00:00
"Written part in {} sec., {} rows, {} uncompressed, {} compressed,"
" {} uncompressed bytes per row, {} compressed bytes per row, compression rate: {}"
" ({} rows/sec., {}/sec. uncompressed, {}/sec. compressed)",
elapsed_seconds,
rows,
2020-05-30 21:35:52 +00:00
ReadableSize(uncompressed_bytes),
ReadableSize(compressed_bytes),
2020-05-23 21:16:05 +00:00
uncompressed_bytes / rows,
compressed_bytes / rows,
uncompressed_bytes / compressed_bytes,
rows / elapsed_seconds,
2020-05-30 21:35:52 +00:00
ReadableSize(uncompressed_bytes / elapsed_seconds),
ReadableSize(compressed_bytes / elapsed_seconds));
}
void Aggregator::writeToTemporaryFile(AggregatedDataVariants & data_variants)
{
return writeToTemporaryFile(data_variants, params.tmp_volume->getNextDisk()->getPath());
}
2015-12-06 14:27:09 +00:00
template <typename Method>
Block Aggregator::convertOneBucketToBlock(
AggregatedDataVariants & data_variants,
Method & method,
bool final,
size_t bucket) const
2015-12-06 14:27:09 +00:00
{
Block block = prepareBlockAndFill(data_variants, final, method.data.impls[bucket].size(),
[bucket, &method, this] (
MutableColumns & key_columns,
AggregateColumnsData & aggregate_columns,
MutableColumns & final_aggregate_columns,
2018-08-27 18:16:32 +00:00
bool final_)
{
convertToBlockImpl(method, method.data.impls[bucket],
2018-08-27 18:16:32 +00:00
key_columns, aggregate_columns, final_aggregate_columns, final_);
});
block.info.bucket_num = bucket;
return block;
2015-12-06 14:27:09 +00:00
}
Block Aggregator::mergeAndConvertOneBucketToBlock(
ManyAggregatedDataVariants & variants,
Arena * arena,
bool final,
2020-01-10 20:24:59 +00:00
size_t bucket,
std::atomic<bool> * is_cancelled) const
{
auto & merged_data = *variants[0];
auto method = merged_data.type;
Block block;
2020-03-08 23:48:08 +00:00
if (false) {} // NOLINT
#define M(NAME) \
else if (method == AggregatedDataVariants::Type::NAME) \
{ \
mergeBucketImpl<decltype(merged_data.NAME)::element_type>(variants, bucket, arena); \
2020-01-10 20:24:59 +00:00
if (is_cancelled && is_cancelled->load(std::memory_order_seq_cst)) \
return {}; \
block = convertOneBucketToBlock(merged_data, *merged_data.NAME, final, bucket); \
}
APPLY_FOR_VARIANTS_TWO_LEVEL(M)
#undef M
return block;
}
2015-12-06 14:27:09 +00:00
template <typename Method>
void Aggregator::writeToTemporaryFileImpl(
AggregatedDataVariants & data_variants,
Method & method,
2017-12-01 21:13:25 +00:00
IBlockOutputStream & out)
{
size_t max_temporary_block_size_rows = 0;
size_t max_temporary_block_size_bytes = 0;
auto update_max_sizes = [&](const Block & block)
{
size_t block_size_rows = block.rows();
size_t block_size_bytes = block.bytes();
if (block_size_rows > max_temporary_block_size_rows)
max_temporary_block_size_rows = block_size_rows;
if (block_size_bytes > max_temporary_block_size_bytes)
max_temporary_block_size_bytes = block_size_bytes;
};
for (size_t bucket = 0; bucket < Method::Data::NUM_BUCKETS; ++bucket)
{
Block block = convertOneBucketToBlock(data_variants, method, false, bucket);
out.write(block);
update_max_sizes(block);
}
if (params.overflow_row)
{
Block block = prepareBlockAndFillWithoutKey(data_variants, false, true);
out.write(block);
update_max_sizes(block);
}
/// Pass ownership of the aggregate functions states:
/// `data_variants` will not destroy them in the destructor, they are now owned by ColumnAggregateFunction objects.
data_variants.aggregator = nullptr;
2020-05-30 21:35:52 +00:00
LOG_TRACE(log, "Max size of temporary block: {} rows, {}.", max_temporary_block_size_rows, ReadableSize(max_temporary_block_size_bytes));
}
bool Aggregator::checkLimits(size_t result_size, bool & no_more_keys) const
{
if (!no_more_keys && params.max_rows_to_group_by && result_size > params.max_rows_to_group_by)
{
switch (params.group_by_overflow_mode)
{
case OverflowMode::THROW:
throw Exception("Limit for rows to GROUP BY exceeded: has " + toString(result_size)
+ " rows, maximum: " + toString(params.max_rows_to_group_by),
2018-03-09 23:23:15 +00:00
ErrorCodes::TOO_MANY_ROWS);
case OverflowMode::BREAK:
return false;
case OverflowMode::ANY:
no_more_keys = true;
break;
}
}
return true;
2014-05-10 05:16:23 +00:00
}
void Aggregator::execute(const BlockInputStreamPtr & stream, AggregatedDataVariants & result)
2011-09-19 01:42:16 +00:00
{
if (isCancelled())
return;
ColumnRawPtrs key_columns(params.keys_size);
AggregateColumns aggregate_columns(params.aggregates_size);
2017-04-02 17:37:49 +00:00
/** Used if there is a limit on the maximum number of rows in the aggregation,
* and if group_by_overflow_mode == ANY.
* In this case, new keys are not added to the set, but aggregation is performed only by
* keys that have already managed to get into the set.
*/
bool no_more_keys = false;
2020-05-23 22:24:01 +00:00
LOG_TRACE(log, "Aggregating");
Stopwatch watch;
size_t src_rows = 0;
size_t src_bytes = 0;
2017-04-02 17:37:49 +00:00
/// Read all the data
while (Block block = stream->read())
{
if (isCancelled())
return;
src_rows += block.rows();
src_bytes += block.bytes();
2019-08-11 21:45:18 +00:00
if (!executeOnBlock(block, result, key_columns, aggregate_columns, no_more_keys))
break;
}
/// If there was no data, and we aggregate without keys, and we must return single row with the result of empty aggregation.
/// To do this, we pass a block with zero rows to aggregate.
if (result.empty() && params.keys_size == 0 && !params.empty_result_for_aggregation_by_empty_set)
2019-08-11 21:45:18 +00:00
executeOnBlock(stream->getHeader(), result, key_columns, aggregate_columns, no_more_keys);
double elapsed_seconds = watch.elapsedSeconds();
size_t rows = result.sizeWithoutOverflowRow();
2020-05-23 21:16:05 +00:00
2020-05-23 22:24:01 +00:00
LOG_TRACE(log, "Aggregated. {} to {} rows (from {}) in {} sec. ({} rows/sec., {}/sec.)",
2020-05-30 21:35:52 +00:00
src_rows, rows, ReadableSize(src_bytes),
2020-05-23 21:16:05 +00:00
elapsed_seconds, src_rows / elapsed_seconds,
2020-05-30 21:35:52 +00:00
ReadableSize(src_bytes / elapsed_seconds));
}
template <typename Method, typename Table>
void Aggregator::convertToBlockImpl(
Method & method,
Table & data,
MutableColumns & key_columns,
AggregateColumnsData & aggregate_columns,
MutableColumns & final_aggregate_columns,
bool final) const
{
if (data.empty())
return;
2015-12-11 00:34:00 +00:00
if (key_columns.size() != params.keys_size)
throw Exception{"Aggregate. Unexpected key columns size.", ErrorCodes::LOGICAL_ERROR};
if (final)
2018-08-27 18:05:28 +00:00
convertToBlockImplFinal(method, data, key_columns, final_aggregate_columns);
else
2018-08-27 18:05:28 +00:00
convertToBlockImplNotFinal(method, data, key_columns, aggregate_columns);
2017-04-02 17:37:49 +00:00
/// In order to release memory early.
data.clearAndShrink();
}
template <typename Mapped>
2020-06-08 12:33:00 +00:00
inline void Aggregator::insertAggregatesIntoColumns(
Mapped & mapped,
MutableColumns & final_aggregate_columns) const
{
/** Final values of aggregate functions are inserted to columns.
* Then states of aggregate functions, that are not longer needed, are destroyed.
*
* We mark already destroyed states with "nullptr" in data,
* so they will not be destroyed in destructor of Aggregator
* (other values will be destroyed in destructor in case of exception).
*
* But it becomes tricky, because we have multiple aggregate states pointed by a single pointer in data.
* So, if exception is thrown in the middle of moving states for different aggregate functions,
* we have to catch exceptions and destroy all the states that are no longer needed,
* to keep the data in consistent state.
*
* It is also tricky, because there are aggregate functions with "-State" modifier.
* When we call "insertResultInto" for them, they insert a pointer to the state to ColumnAggregateFunction
* and ColumnAggregateFunction will take ownership of this state.
* So, for aggregate functions with "-State" modifier, the state must not be destroyed
* after it has been transferred to ColumnAggregateFunction.
* But we should mark that the data no longer owns these states.
*/
size_t insert_i = 0;
std::exception_ptr exception;
try
{
/// Insert final values of aggregate functions into columns.
for (; insert_i < params.aggregates_size; ++insert_i)
aggregate_functions[insert_i]->insertResultInto(
mapped + offsets_of_aggregate_states[insert_i],
*final_aggregate_columns[insert_i]);
}
catch (...)
{
exception = std::current_exception();
}
/** Destroy states that are no longer needed. This loop does not throw.
*
* Don't destroy states for "-State" aggregate functions,
* because the ownership of this state is transferred to ColumnAggregateFunction
* and ColumnAggregateFunction will take care.
*
* But it's only for states that has been transferred to ColumnAggregateFunction
* before exception has been thrown;
*/
for (size_t destroy_i = 0; destroy_i < params.aggregates_size; ++destroy_i)
{
/// If ownership was not transferred to ColumnAggregateFunction.
if (!(destroy_i < insert_i && aggregate_functions[destroy_i]->isState()))
aggregate_functions[destroy_i]->destroy(
mapped + offsets_of_aggregate_states[destroy_i]);
}
/// Mark the cell as destroyed so it will not be destroyed in destructor.
mapped = nullptr;
if (exception)
std::rethrow_exception(exception);
}
template <typename Method, typename Table>
void NO_INLINE Aggregator::convertToBlockImplFinal(
Method & method,
Table & data,
MutableColumns & key_columns,
2018-08-27 18:05:28 +00:00
MutableColumns & final_aggregate_columns) const
{
if constexpr (Method::low_cardinality_optimization)
{
if (data.hasNullKeyData())
{
2019-02-08 16:54:04 +00:00
key_columns[0]->insertDefault();
insertAggregatesIntoColumns(data.getNullKeyData(), final_aggregate_columns);
}
}
data.forEachValue([&](const auto & key, auto & mapped)
{
method.insertKeyIntoColumns(key, key_columns, key_sizes);
insertAggregatesIntoColumns(mapped, final_aggregate_columns);
2020-06-07 22:46:58 +00:00
});
}
template <typename Method, typename Table>
void NO_INLINE Aggregator::convertToBlockImplNotFinal(
Method & method,
Table & data,
MutableColumns & key_columns,
2018-08-27 18:05:28 +00:00
AggregateColumnsData & aggregate_columns) const
{
if constexpr (Method::low_cardinality_optimization)
{
if (data.hasNullKeyData())
{
2019-02-08 16:54:04 +00:00
key_columns[0]->insertDefault();
for (size_t i = 0; i < params.aggregates_size; ++i)
aggregate_columns[i]->push_back(data.getNullKeyData() + offsets_of_aggregate_states[i]);
2020-06-07 22:46:58 +00:00
data.getNullKeyData() = nullptr;
}
}
data.forEachValue([&](const auto & key, auto & mapped)
{
method.insertKeyIntoColumns(key, key_columns, key_sizes);
2017-04-02 17:37:49 +00:00
/// reserved, so push_back does not throw exceptions
for (size_t i = 0; i < params.aggregates_size; ++i)
aggregate_columns[i]->push_back(mapped + offsets_of_aggregate_states[i]);
2015-12-23 07:29:20 +00:00
mapped = nullptr;
});
2011-09-19 01:42:16 +00:00
}
template <typename Filler>
Block Aggregator::prepareBlockAndFill(
AggregatedDataVariants & data_variants,
bool final,
size_t rows,
Filler && filler) const
2012-02-27 06:28:20 +00:00
{
MutableColumns key_columns(params.keys_size);
MutableColumns aggregate_columns(params.aggregates_size);
MutableColumns final_aggregate_columns(params.aggregates_size);
AggregateColumnsData aggregate_columns_data(params.aggregates_size);
Block header = getHeader(final);
for (size_t i = 0; i < params.keys_size; ++i)
{
key_columns[i] = header.safeGetByPosition(i).type->createColumn();
key_columns[i]->reserve(rows);
}
for (size_t i = 0; i < params.aggregates_size; ++i)
{
if (!final)
{
const auto & aggregate_column_name = params.aggregates[i].column_name;
aggregate_columns[i] = header.getByName(aggregate_column_name).type->createColumn();
2017-04-02 17:37:49 +00:00
/// The ColumnAggregateFunction column captures the shared ownership of the arena with the aggregate function states.
ColumnAggregateFunction & column_aggregate_func = assert_cast<ColumnAggregateFunction &>(*aggregate_columns[i]);
2020-03-08 23:48:08 +00:00
for (auto & pool : data_variants.aggregates_pools)
column_aggregate_func.addArena(pool);
aggregate_columns_data[i] = &column_aggregate_func.getData();
aggregate_columns_data[i]->reserve(rows);
}
else
{
final_aggregate_columns[i] = aggregate_functions[i]->getReturnType()->createColumn();
final_aggregate_columns[i]->reserve(rows);
if (aggregate_functions[i]->isState())
{
2017-04-02 17:37:49 +00:00
/// The ColumnAggregateFunction column captures the shared ownership of the arena with aggregate function states.
ColumnAggregateFunction & column_aggregate_func = assert_cast<ColumnAggregateFunction &>(*final_aggregate_columns[i]);
2020-03-08 23:48:08 +00:00
for (auto & pool : data_variants.aggregates_pools)
column_aggregate_func.addArena(pool);
}
}
}
2018-08-27 18:05:28 +00:00
filler(key_columns, aggregate_columns_data, final_aggregate_columns, final);
Block res = header.cloneEmpty();
for (size_t i = 0; i < params.keys_size; ++i)
res.getByPosition(i).column = std::move(key_columns[i]);
for (size_t i = 0; i < params.aggregates_size; ++i)
{
const auto & aggregate_column_name = params.aggregates[i].column_name;
if (final)
res.getByName(aggregate_column_name).column = std::move(final_aggregate_columns[i]);
else
res.getByName(aggregate_column_name).column = std::move(aggregate_columns[i]);
}
2017-04-02 17:37:49 +00:00
/// Change the size of the columns-constants in the block.
size_t columns = header.columns();
for (size_t i = 0; i < columns; ++i)
if (isColumnConst(*res.getByPosition(i).column))
res.getByPosition(i).column = res.getByPosition(i).column->cut(0, rows);
return res;
}
2020-04-18 09:51:21 +00:00
void Aggregator::fillAggregateColumnsWithSingleKey(
AggregatedDataVariants & data_variants,
MutableColumns & final_aggregate_columns)
{
AggregatedDataWithoutKey & data = data_variants.without_key;
2014-05-28 14:54:42 +00:00
2020-04-18 09:51:21 +00:00
for (size_t i = 0; i < params.aggregates_size; ++i)
{
2020-05-12 14:50:13 +00:00
ColumnAggregateFunction & column_aggregate_func = assert_cast<ColumnAggregateFunction &>(*final_aggregate_columns[i]);
for (auto & pool : data_variants.aggregates_pools)
{
column_aggregate_func.addArena(pool);
}
2020-05-14 14:20:49 +00:00
column_aggregate_func.getData().push_back(data + offsets_of_aggregate_states[i]);
2020-04-18 09:51:21 +00:00
}
2020-05-12 14:50:13 +00:00
data = nullptr;
2020-04-18 09:51:21 +00:00
}
2014-05-28 14:54:42 +00:00
2020-04-18 09:51:21 +00:00
void Aggregator::createStatesAndFillKeyColumnsWithSingleKey(
AggregatedDataVariants & data_variants,
2020-05-08 13:28:18 +00:00
Columns & key_columns,
2020-04-18 09:51:21 +00:00
size_t key_row,
MutableColumns & final_key_columns)
{
AggregateDataPtr place = data_variants.aggregates_pool->alignedAlloc(total_size_of_aggregate_states, align_aggregate_states);
createAggregateStates(place);
data_variants.without_key = place;
for (size_t i = 0; i < params.keys_size; ++i)
{
2020-05-07 20:13:51 +00:00
final_key_columns[i]->insertFrom(*key_columns[i].get(), key_row);
2020-04-18 09:51:21 +00:00
}
}
Block Aggregator::prepareBlockAndFillWithoutKey(AggregatedDataVariants & data_variants, bool final, bool is_overflows) const
{
size_t rows = 1;
auto filler = [&data_variants, this](
MutableColumns & key_columns,
AggregateColumnsData & aggregate_columns,
MutableColumns & final_aggregate_columns,
2018-08-27 18:16:32 +00:00
bool final_)
{
if (data_variants.type == AggregatedDataVariants::Type::without_key || params.overflow_row)
{
AggregatedDataWithoutKey & data = data_variants.without_key;
if (!final_)
{
for (size_t i = 0; i < params.aggregates_size; ++i)
aggregate_columns[i]->push_back(data + offsets_of_aggregate_states[i]);
data = nullptr;
}
else
{
insertAggregatesIntoColumns(data, final_aggregate_columns);
}
if (params.overflow_row)
for (size_t i = 0; i < params.keys_size; ++i)
key_columns[i]->insertDefault();
}
};
Block block = prepareBlockAndFill(data_variants, final, rows, filler);
if (is_overflows)
block.info.is_overflows = true;
if (final)
destroyWithoutKey(data_variants);
return block;
}
Block Aggregator::prepareBlockAndFillSingleLevel(AggregatedDataVariants & data_variants, bool final) const
{
size_t rows = data_variants.sizeWithoutOverflowRow();
auto filler = [&data_variants, this](
MutableColumns & key_columns,
AggregateColumnsData & aggregate_columns,
MutableColumns & final_aggregate_columns,
2018-08-27 18:16:32 +00:00
bool final_)
{
#define M(NAME) \
else if (data_variants.type == AggregatedDataVariants::Type::NAME) \
convertToBlockImpl(*data_variants.NAME, data_variants.NAME->data, \
2018-08-27 18:16:32 +00:00
key_columns, aggregate_columns, final_aggregate_columns, final_);
2020-03-08 23:48:08 +00:00
if (false) {} // NOLINT
APPLY_FOR_VARIANTS_SINGLE_LEVEL(M)
#undef M
else
throw Exception("Unknown aggregated data variant.", ErrorCodes::UNKNOWN_AGGREGATED_DATA_VARIANT);
};
return prepareBlockAndFill(data_variants, final, rows, filler);
}
BlocksList Aggregator::prepareBlocksAndFillTwoLevel(AggregatedDataVariants & data_variants, bool final, ThreadPool * thread_pool) const
{
#define M(NAME) \
else if (data_variants.type == AggregatedDataVariants::Type::NAME) \
return prepareBlocksAndFillTwoLevelImpl(data_variants, *data_variants.NAME, final, thread_pool);
2020-03-08 23:48:08 +00:00
if (false) {} // NOLINT
APPLY_FOR_VARIANTS_TWO_LEVEL(M)
#undef M
else
throw Exception("Unknown aggregated data variant.", ErrorCodes::UNKNOWN_AGGREGATED_DATA_VARIANT);
}
template <typename Method>
BlocksList Aggregator::prepareBlocksAndFillTwoLevelImpl(
AggregatedDataVariants & data_variants,
Method & method,
bool final,
ThreadPool * thread_pool) const
{
auto converter = [&](size_t bucket, ThreadGroupStatusPtr thread_group)
{
if (thread_group)
CurrentThread::attachToIfDetached(thread_group);
return convertOneBucketToBlock(data_variants, method, final, bucket);
};
2017-04-02 17:37:49 +00:00
/// packaged_task is used to ensure that exceptions are automatically thrown into the main stream.
std::vector<std::packaged_task<Block()>> tasks(Method::Data::NUM_BUCKETS);
try
{
for (size_t bucket = 0; bucket < Method::Data::NUM_BUCKETS; ++bucket)
{
if (method.data.impls[bucket].empty())
continue;
2020-04-22 05:39:31 +00:00
tasks[bucket] = std::packaged_task<Block()>([group = CurrentThread::getGroup(), bucket, &converter]{ return converter(bucket, group); });
if (thread_pool)
thread_pool->scheduleOrThrowOnError([bucket, &tasks] { tasks[bucket](); });
else
tasks[bucket]();
}
}
catch (...)
{
2017-04-02 17:37:49 +00:00
/// If this is not done, then in case of an exception, tasks will be destroyed before the threads are completed, and it will be bad.
if (thread_pool)
thread_pool->wait();
throw;
}
if (thread_pool)
thread_pool->wait();
BlocksList blocks;
for (auto & task : tasks)
{
if (!task.valid())
continue;
blocks.emplace_back(task.get_future().get());
}
return blocks;
}
2015-12-08 02:01:46 +00:00
BlocksList Aggregator::convertToBlocks(AggregatedDataVariants & data_variants, bool final, size_t max_threads) const
{
if (isCancelled())
return BlocksList();
2015-04-16 14:27:56 +00:00
2020-05-23 22:24:01 +00:00
LOG_TRACE(log, "Converting aggregated data to blocks");
Stopwatch watch;
BlocksList blocks;
2017-04-02 17:37:49 +00:00
/// In what data structure is the data aggregated?
if (data_variants.empty())
return blocks;
std::unique_ptr<ThreadPool> thread_pool;
2017-04-02 17:37:49 +00:00
if (max_threads > 1 && data_variants.sizeWithoutOverflowRow() > 100000 /// TODO Make a custom threshold.
&& data_variants.isTwoLevel()) /// TODO Use the shared thread pool with the `merge` function.
thread_pool = std::make_unique<ThreadPool>(max_threads);
if (isCancelled())
return BlocksList();
2015-04-16 14:27:56 +00:00
if (data_variants.without_key)
blocks.emplace_back(prepareBlockAndFillWithoutKey(
data_variants, final, data_variants.type != AggregatedDataVariants::Type::without_key));
if (isCancelled())
return BlocksList();
2015-04-16 14:27:56 +00:00
if (data_variants.type != AggregatedDataVariants::Type::without_key)
{
if (!data_variants.isTwoLevel())
blocks.emplace_back(prepareBlockAndFillSingleLevel(data_variants, final));
else
blocks.splice(blocks.end(), prepareBlocksAndFillTwoLevel(data_variants, final, thread_pool.get()));
}
if (!final)
{
2017-04-02 17:37:49 +00:00
/// data_variants will not destroy the states of aggregate functions in the destructor.
/// Now ColumnAggregateFunction owns the states.
data_variants.aggregator = nullptr;
}
if (isCancelled())
return BlocksList();
2015-04-16 14:27:56 +00:00
size_t rows = 0;
size_t bytes = 0;
for (const auto & block : blocks)
{
rows += block.rows();
bytes += block.bytes();
}
2012-02-27 06:28:20 +00:00
double elapsed_seconds = watch.elapsedSeconds();
2020-05-23 22:24:01 +00:00
LOG_TRACE(log,
2020-05-23 21:26:45 +00:00
"Converted aggregated data to blocks. {} rows, {} in {} sec. ({} rows/sec., {}/sec.)",
2020-05-30 21:35:52 +00:00
rows, ReadableSize(bytes),
2020-05-23 21:26:45 +00:00
elapsed_seconds, rows / elapsed_seconds,
2020-05-30 21:35:52 +00:00
ReadableSize(bytes / elapsed_seconds));
2012-05-31 00:33:42 +00:00
return blocks;
2012-02-27 06:28:20 +00:00
}
template <typename Method, typename Table>
void NO_INLINE Aggregator::mergeDataNullKey(
Table & table_dst,
Table & table_src,
Arena * arena) const
{
if constexpr (Method::low_cardinality_optimization)
{
if (table_src.hasNullKeyData())
{
if (!table_dst.hasNullKeyData())
{
table_dst.hasNullKeyData() = true;
table_dst.getNullKeyData() = table_src.getNullKeyData();
}
else
{
for (size_t i = 0; i < params.aggregates_size; ++i)
aggregate_functions[i]->merge(
table_dst.getNullKeyData() + offsets_of_aggregate_states[i],
table_src.getNullKeyData() + offsets_of_aggregate_states[i],
arena);
for (size_t i = 0; i < params.aggregates_size; ++i)
aggregate_functions[i]->destroy(
table_src.getNullKeyData() + offsets_of_aggregate_states[i]);
}
table_src.hasNullKeyData() = false;
table_src.getNullKeyData() = nullptr;
}
}
}
template <typename Method, typename Table>
void NO_INLINE Aggregator::mergeDataImpl(
Table & table_dst,
Table & table_src,
Arena * arena) const
{
if constexpr (Method::low_cardinality_optimization)
mergeDataNullKey<Method, Table>(table_dst, table_src, arena);
table_src.mergeToViaEmplace(table_dst,
[&](AggregateDataPtr & dst, AggregateDataPtr & src, bool inserted)
{
if (!inserted)
{
for (size_t i = 0; i < params.aggregates_size; ++i)
aggregate_functions[i]->merge(
dst + offsets_of_aggregate_states[i],
src + offsets_of_aggregate_states[i],
arena);
for (size_t i = 0; i < params.aggregates_size; ++i)
aggregate_functions[i]->destroy(src + offsets_of_aggregate_states[i]);
}
else
{
dst = src;
}
src = nullptr;
});
table_src.clearAndShrink();
}
template <typename Method, typename Table>
void NO_INLINE Aggregator::mergeDataNoMoreKeysImpl(
Table & table_dst,
AggregatedDataWithoutKey & overflows,
Table & table_src,
Arena * arena) const
{
/// Note : will create data for NULL key if not exist
if constexpr (Method::low_cardinality_optimization)
mergeDataNullKey<Method, Table>(table_dst, table_src, arena);
table_src.mergeToViaFind(table_dst, [&](AggregateDataPtr dst, AggregateDataPtr & src, bool found)
{
AggregateDataPtr res_data = found ? dst : overflows;
for (size_t i = 0; i < params.aggregates_size; ++i)
aggregate_functions[i]->merge(
res_data + offsets_of_aggregate_states[i],
src + offsets_of_aggregate_states[i],
arena);
for (size_t i = 0; i < params.aggregates_size; ++i)
aggregate_functions[i]->destroy(src + offsets_of_aggregate_states[i]);
2015-12-23 07:06:34 +00:00
src = nullptr;
});
table_src.clearAndShrink();
}
template <typename Method, typename Table>
void NO_INLINE Aggregator::mergeDataOnlyExistingKeysImpl(
Table & table_dst,
Table & table_src,
Arena * arena) const
{
/// Note : will create data for NULL key if not exist
if constexpr (Method::low_cardinality_optimization)
mergeDataNullKey<Method, Table>(table_dst, table_src, arena);
table_src.mergeToViaFind(table_dst,
[&](AggregateDataPtr dst, AggregateDataPtr & src, bool found)
{
if (!found)
return;
for (size_t i = 0; i < params.aggregates_size; ++i)
aggregate_functions[i]->merge(
dst + offsets_of_aggregate_states[i],
src + offsets_of_aggregate_states[i],
arena);
for (size_t i = 0; i < params.aggregates_size; ++i)
aggregate_functions[i]->destroy(src + offsets_of_aggregate_states[i]);
2015-12-23 07:06:34 +00:00
src = nullptr;
});
table_src.clearAndShrink();
}
void NO_INLINE Aggregator::mergeWithoutKeyDataImpl(
ManyAggregatedDataVariants & non_empty_data) const
{
AggregatedDataVariantsPtr & res = non_empty_data[0];
2018-08-27 18:16:32 +00:00
/// We merge all aggregation results to the first.
for (size_t result_num = 1, size = non_empty_data.size(); result_num < size; ++result_num)
{
AggregatedDataWithoutKey & res_data = res->without_key;
2018-08-27 18:16:32 +00:00
AggregatedDataWithoutKey & current_data = non_empty_data[result_num]->without_key;
for (size_t i = 0; i < params.aggregates_size; ++i)
aggregate_functions[i]->merge(res_data + offsets_of_aggregate_states[i], current_data + offsets_of_aggregate_states[i], res->aggregates_pool);
for (size_t i = 0; i < params.aggregates_size; ++i)
aggregate_functions[i]->destroy(current_data + offsets_of_aggregate_states[i]);
current_data = nullptr;
}
}
template <typename Method>
void NO_INLINE Aggregator::mergeSingleLevelDataImpl(
ManyAggregatedDataVariants & non_empty_data) const
{
AggregatedDataVariantsPtr & res = non_empty_data[0];
bool no_more_keys = false;
2018-08-27 18:16:32 +00:00
/// We merge all aggregation results to the first.
for (size_t result_num = 1, size = non_empty_data.size(); result_num < size; ++result_num)
{
if (!checkLimits(res->sizeWithoutOverflowRow(), no_more_keys))
break;
2018-08-27 18:16:32 +00:00
AggregatedDataVariants & current = *non_empty_data[result_num];
if (!no_more_keys)
mergeDataImpl<Method>(
getDataVariant<Method>(*res).data,
getDataVariant<Method>(current).data,
res->aggregates_pool);
else if (res->without_key)
mergeDataNoMoreKeysImpl<Method>(
getDataVariant<Method>(*res).data,
res->without_key,
getDataVariant<Method>(current).data,
res->aggregates_pool);
else
mergeDataOnlyExistingKeysImpl<Method>(
getDataVariant<Method>(*res).data,
getDataVariant<Method>(current).data,
res->aggregates_pool);
2017-04-02 17:37:49 +00:00
/// `current` will not destroy the states of aggregate functions in the destructor
current.aggregator = nullptr;
}
}
2020-05-08 20:55:08 +00:00
#define M(NAME) \
template void NO_INLINE Aggregator::mergeSingleLevelDataImpl<decltype(AggregatedDataVariants::NAME)::element_type>( \
ManyAggregatedDataVariants & non_empty_data) const;
APPLY_FOR_VARIANTS_SINGLE_LEVEL(M)
#undef M
template <typename Method>
void NO_INLINE Aggregator::mergeBucketImpl(
2020-01-10 20:24:59 +00:00
ManyAggregatedDataVariants & data, Int32 bucket, Arena * arena, std::atomic<bool> * is_cancelled) const
{
2018-08-27 18:16:32 +00:00
/// We merge all aggregation results to the first.
AggregatedDataVariantsPtr & res = data[0];
2018-08-27 18:16:32 +00:00
for (size_t result_num = 1, size = data.size(); result_num < size; ++result_num)
{
2020-01-10 20:24:59 +00:00
if (is_cancelled && is_cancelled->load(std::memory_order_seq_cst))
return;
2018-08-27 18:16:32 +00:00
AggregatedDataVariants & current = *data[result_num];
mergeDataImpl<Method>(
getDataVariant<Method>(*res).data.impls[bucket],
getDataVariant<Method>(current).data.impls[bucket],
arena);
}
}
2017-04-02 17:37:49 +00:00
/** Combines aggregation states together, turns them into blocks, and outputs streams.
* If the aggregation states are two-level, then it produces blocks strictly in order of 'bucket_num'.
2017-04-02 17:37:49 +00:00
* (This is important for distributed processing.)
* In doing so, it can handle different buckets in parallel, using up to `threads` threads.
*/
class MergingAndConvertingBlockInputStream : public IBlockInputStream
{
public:
2017-04-02 17:37:49 +00:00
/** The input is a set of non-empty sets of partially aggregated data,
* which are all either single-level, or are two-level.
*/
MergingAndConvertingBlockInputStream(const Aggregator & aggregator_, ManyAggregatedDataVariants & data_, bool final_, size_t threads_)
: aggregator(aggregator_), data(data_), final(final_), threads(threads_)
{
/// At least we need one arena in first data item per thread
if (!data.empty() && threads > data[0]->aggregates_pools.size())
{
Arenas & first_pool = data[0]->aggregates_pools;
for (size_t j = first_pool.size(); j < threads; j++)
first_pool.emplace_back(std::make_shared<Arena>());
}
}
String getName() const override { return "MergingAndConverting"; }
Block getHeader() const override { return aggregator.getHeader(final); }
~MergingAndConvertingBlockInputStream() override
{
2020-05-30 21:57:37 +00:00
LOG_TRACE(&Poco::Logger::get(__PRETTY_FUNCTION__), "Waiting for threads to finish");
/// We need to wait for threads to finish before destructor of 'parallel_merge_data',
/// because the threads access 'parallel_merge_data'.
if (parallel_merge_data)
parallel_merge_data->pool.wait();
}
protected:
Block readImpl() override
{
if (data.empty())
return {};
if (current_bucket_num >= NUM_BUCKETS)
return {};
AggregatedDataVariantsPtr & first = data[0];
if (current_bucket_num == -1)
{
++current_bucket_num;
if (first->type == AggregatedDataVariants::Type::without_key || aggregator.params.overflow_row)
{
aggregator.mergeWithoutKeyDataImpl(data);
return aggregator.prepareBlockAndFillWithoutKey(
*first, final, first->type != AggregatedDataVariants::Type::without_key);
}
}
if (!first->isTwoLevel())
{
if (current_bucket_num > 0)
return {};
if (first->type == AggregatedDataVariants::Type::without_key)
return {};
++current_bucket_num;
#define M(NAME) \
else if (first->type == AggregatedDataVariants::Type::NAME) \
aggregator.mergeSingleLevelDataImpl<decltype(first->NAME)::element_type>(data);
2020-03-08 23:48:08 +00:00
if (false) {} // NOLINT
APPLY_FOR_VARIANTS_SINGLE_LEVEL(M)
#undef M
else
throw Exception("Unknown aggregated data variant.", ErrorCodes::UNKNOWN_AGGREGATED_DATA_VARIANT);
return aggregator.prepareBlockAndFillSingleLevel(*first, final);
}
else
{
if (!parallel_merge_data)
{
parallel_merge_data = std::make_unique<ParallelMergeData>(threads);
for (size_t i = 0; i < threads; ++i)
scheduleThreadForNextBucket();
}
Block res;
while (true)
{
2019-01-02 06:44:36 +00:00
std::unique_lock lock(parallel_merge_data->mutex);
if (parallel_merge_data->exception)
std::rethrow_exception(parallel_merge_data->exception);
auto it = parallel_merge_data->ready_blocks.find(current_bucket_num);
if (it != parallel_merge_data->ready_blocks.end())
{
++current_bucket_num;
scheduleThreadForNextBucket();
if (it->second)
{
res.swap(it->second);
break;
}
else if (current_bucket_num >= NUM_BUCKETS)
break;
}
parallel_merge_data->condvar.wait(lock);
}
return res;
}
}
private:
const Aggregator & aggregator;
ManyAggregatedDataVariants data;
bool final;
size_t threads;
Int32 current_bucket_num = -1;
Int32 max_scheduled_bucket_num = -1;
static constexpr Int32 NUM_BUCKETS = 256;
struct ParallelMergeData
{
std::map<Int32, Block> ready_blocks;
std::exception_ptr exception;
std::mutex mutex;
std::condition_variable condvar;
ThreadPool pool;
2019-08-03 11:02:40 +00:00
explicit ParallelMergeData(size_t threads_) : pool(threads_) {}
};
std::unique_ptr<ParallelMergeData> parallel_merge_data;
void scheduleThreadForNextBucket()
{
++max_scheduled_bucket_num;
if (max_scheduled_bucket_num >= NUM_BUCKETS)
return;
2020-04-22 05:39:31 +00:00
parallel_merge_data->pool.scheduleOrThrowOnError(
2020-04-22 18:41:19 +00:00
[this, max_scheduled_bucket_num = max_scheduled_bucket_num, group = CurrentThread::getGroup()]
{ return thread(max_scheduled_bucket_num, group); });
}
void thread(Int32 bucket_num, ThreadGroupStatusPtr thread_group)
{
try
{
setThreadName("MergingAggregtd");
if (thread_group)
CurrentThread::attachToIfDetached(thread_group);
CurrentMetrics::Increment metric_increment{CurrentMetrics::QueryThread};
/// TODO: add no_more_keys support maybe
auto & merged_data = *data[0];
auto method = merged_data.type;
Block block;
/// Select Arena to avoid race conditions
size_t thread_number = static_cast<size_t>(bucket_num) % threads;
Arena * arena = merged_data.aggregates_pools.at(thread_number).get();
2020-03-08 23:48:08 +00:00
if (false) {} // NOLINT
#define M(NAME) \
else if (method == AggregatedDataVariants::Type::NAME) \
{ \
aggregator.mergeBucketImpl<decltype(merged_data.NAME)::element_type>(data, bucket_num, arena); \
block = aggregator.convertOneBucketToBlock(merged_data, *merged_data.NAME, final, bucket_num); \
}
APPLY_FOR_VARIANTS_TWO_LEVEL(M)
#undef M
2019-01-02 06:44:36 +00:00
std::lock_guard lock(parallel_merge_data->mutex);
parallel_merge_data->ready_blocks[bucket_num] = std::move(block);
}
catch (...)
{
2019-01-02 06:44:36 +00:00
std::lock_guard lock(parallel_merge_data->mutex);
if (!parallel_merge_data->exception)
parallel_merge_data->exception = std::current_exception();
}
parallel_merge_data->condvar.notify_all();
}
};
ManyAggregatedDataVariants Aggregator::prepareVariantsToMerge(ManyAggregatedDataVariants & data_variants) const
{
if (data_variants.empty())
throw Exception("Empty data passed to Aggregator::mergeAndConvertToBlocks.", ErrorCodes::EMPTY_DATA_PASSED);
2020-05-23 22:24:01 +00:00
LOG_TRACE(log, "Merging aggregated data");
ManyAggregatedDataVariants non_empty_data;
non_empty_data.reserve(data_variants.size());
for (auto & data : data_variants)
if (!data->empty())
non_empty_data.push_back(data);
if (non_empty_data.empty())
return {};
if (non_empty_data.size() > 1)
{
2017-04-02 17:37:49 +00:00
/// Sort the states in descending order so that the merge is more efficient (since all states are merged into the first).
std::sort(non_empty_data.begin(), non_empty_data.end(),
[](const AggregatedDataVariantsPtr & lhs, const AggregatedDataVariantsPtr & rhs)
{
return lhs->sizeWithoutOverflowRow() > rhs->sizeWithoutOverflowRow();
});
}
2017-04-02 17:37:49 +00:00
/// If at least one of the options is two-level, then convert all the options into two-level ones, if there are not such.
/// Note - perhaps it would be more optimal not to convert single-level versions before the merge, but merge them separately, at the end.
bool has_at_least_one_two_level = false;
for (const auto & variant : non_empty_data)
{
if (variant->isTwoLevel())
{
has_at_least_one_two_level = true;
break;
}
}
if (has_at_least_one_two_level)
for (auto & variant : non_empty_data)
if (!variant->isTwoLevel())
variant->convertToTwoLevel();
AggregatedDataVariantsPtr & first = non_empty_data[0];
for (size_t i = 1, size = non_empty_data.size(); i < size; ++i)
{
if (first->type != non_empty_data[i]->type)
throw Exception("Cannot merge different aggregated data variants.", ErrorCodes::CANNOT_MERGE_DIFFERENT_AGGREGATED_DATA_VARIANTS);
2017-04-02 17:37:49 +00:00
/** Elements from the remaining sets can be moved to the first data set.
* Therefore, it must own all the arenas of all other sets.
*/
first->aggregates_pools.insert(first->aggregates_pools.end(),
non_empty_data[i]->aggregates_pools.begin(), non_empty_data[i]->aggregates_pools.end());
}
return non_empty_data;
}
std::unique_ptr<IBlockInputStream> Aggregator::mergeAndConvertToBlocks(
ManyAggregatedDataVariants & data_variants, bool final, size_t max_threads) const
{
ManyAggregatedDataVariants non_empty_data = prepareVariantsToMerge(data_variants);
if (non_empty_data.empty())
return std::make_unique<NullBlockInputStream>(getHeader(final));
return std::make_unique<MergingAndConvertingBlockInputStream>(*this, non_empty_data, final, max_threads);
}
template <bool no_more_keys, typename Method, typename Table>
void NO_INLINE Aggregator::mergeStreamsImplCase(
Block & block,
Arena * aggregates_pool,
2019-01-21 10:39:24 +00:00
Method & method [[maybe_unused]],
Table & data,
AggregateDataPtr overflow_row) const
{
ColumnRawPtrs key_columns(params.keys_size);
AggregateColumnsConstData aggregate_columns(params.aggregates_size);
2017-04-02 17:37:49 +00:00
/// Remember the columns we will work with
for (size_t i = 0; i < params.keys_size; ++i)
key_columns[i] = block.safeGetByPosition(i).column.get();
for (size_t i = 0; i < params.aggregates_size; ++i)
{
const auto & aggregate_column_name = params.aggregates[i].column_name;
aggregate_columns[i] = &typeid_cast<const ColumnAggregateFunction &>(*block.getByName(aggregate_column_name).column).getData();
}
2019-01-21 10:39:24 +00:00
typename Method::State state(key_columns, key_sizes, aggregation_state_cache);
2017-04-02 17:37:49 +00:00
/// For all rows.
size_t rows = block.rows();
for (size_t i = 0; i < rows; ++i)
{
2019-01-21 10:39:53 +00:00
AggregateDataPtr aggregate_data = nullptr;
if (!no_more_keys)
2019-01-21 10:39:53 +00:00
{
auto emplace_result = state.emplaceKey(data, i, *aggregates_pool);
if (emplace_result.isInserted())
{
emplace_result.setMapped(nullptr);
aggregate_data = aggregates_pool->alignedAlloc(total_size_of_aggregate_states, align_aggregate_states);
createAggregateStates(aggregate_data);
emplace_result.setMapped(aggregate_data);
}
else
aggregate_data = emplace_result.getMapped();
}
else
{
2019-01-21 10:39:53 +00:00
auto find_result = state.findKey(data, i, *aggregates_pool);
if (find_result.isFound())
aggregate_data = find_result.getMapped();
}
/// aggregate_date == nullptr means that the new key did not fit in the hash table because of no_more_keys.
2017-04-02 17:37:49 +00:00
/// If the key does not fit, and the data does not need to be aggregated into a separate row, then there's nothing to do.
if (!aggregate_data && !overflow_row)
continue;
2019-01-21 10:39:53 +00:00
AggregateDataPtr value = aggregate_data ? aggregate_data : overflow_row;
2017-04-02 17:37:49 +00:00
/// Merge state of aggregate functions.
for (size_t j = 0; j < params.aggregates_size; ++j)
aggregate_functions[j]->merge(
value + offsets_of_aggregate_states[j],
(*aggregate_columns[j])[i],
aggregates_pool);
}
2017-04-02 17:37:49 +00:00
/// Early release memory.
block.clear();
}
template <typename Method, typename Table>
void NO_INLINE Aggregator::mergeStreamsImpl(
Block & block,
Arena * aggregates_pool,
Method & method,
Table & data,
AggregateDataPtr overflow_row,
bool no_more_keys) const
{
if (!no_more_keys)
2018-08-27 18:05:28 +00:00
mergeStreamsImplCase<false>(block, aggregates_pool, method, data, overflow_row);
else
2018-08-27 18:05:28 +00:00
mergeStreamsImplCase<true>(block, aggregates_pool, method, data, overflow_row);
}
void NO_INLINE Aggregator::mergeWithoutKeyStreamsImpl(
Block & block,
AggregatedDataVariants & result) const
{
AggregateColumnsConstData aggregate_columns(params.aggregates_size);
2017-04-02 17:37:49 +00:00
/// Remember the columns we will work with
for (size_t i = 0; i < params.aggregates_size; ++i)
{
const auto & aggregate_column_name = params.aggregates[i].column_name;
aggregate_columns[i] = &typeid_cast<const ColumnAggregateFunction &>(*block.getByName(aggregate_column_name).column).getData();
}
AggregatedDataWithoutKey & res = result.without_key;
if (!res)
{
2018-09-01 03:17:43 +00:00
AggregateDataPtr place = result.aggregates_pool->alignedAlloc(total_size_of_aggregate_states, align_aggregate_states);
createAggregateStates(place);
res = place;
}
2017-04-02 17:37:49 +00:00
/// Adding Values
for (size_t i = 0; i < params.aggregates_size; ++i)
aggregate_functions[i]->merge(res + offsets_of_aggregate_states[i], (*aggregate_columns[i])[0], result.aggregates_pool);
2017-04-02 17:37:49 +00:00
/// Early release memory.
block.clear();
}
void Aggregator::mergeStream(const BlockInputStreamPtr & stream, AggregatedDataVariants & result, size_t max_threads)
2012-05-30 01:38:02 +00:00
{
if (isCancelled())
return;
2015-04-16 14:27:56 +00:00
2017-04-02 17:37:49 +00:00
/** If the remote servers used a two-level aggregation method,
* then blocks will contain information about the number of the bucket.
* Then the calculations can be parallelized by buckets.
* We decompose the blocks to the bucket numbers indicated in them.
*/
BucketToBlocks bucket_to_blocks;
2017-04-02 17:37:49 +00:00
/// Read all the data.
2020-05-23 22:24:01 +00:00
LOG_TRACE(log, "Reading blocks of partially aggregated data.");
size_t total_input_rows = 0;
size_t total_input_blocks = 0;
while (Block block = stream->read())
{
if (isCancelled())
return;
2015-04-16 14:27:56 +00:00
total_input_rows += block.rows();
++total_input_blocks;
bucket_to_blocks[block.info.bucket_num].emplace_back(std::move(block));
}
2020-05-23 22:24:01 +00:00
LOG_TRACE(log, "Read {} blocks of partially aggregated data, total {} rows.", total_input_blocks, total_input_rows);
2019-03-04 16:06:28 +00:00
mergeBlocks(bucket_to_blocks, result, max_threads);
}
void Aggregator::mergeBlocks(BucketToBlocks bucket_to_blocks, AggregatedDataVariants & result, size_t max_threads)
{
if (bucket_to_blocks.empty())
return;
2019-03-15 17:06:32 +00:00
UInt64 total_input_rows = 0;
2019-03-04 16:06:28 +00:00
for (auto & bucket : bucket_to_blocks)
for (auto & block : bucket.second)
2019-03-15 17:06:32 +00:00
total_input_rows += block.rows();
2019-03-04 16:06:28 +00:00
2017-04-02 17:37:49 +00:00
/** `minus one` means the absence of information about the bucket
* - in the case of single-level aggregation, as well as for blocks with "overflowing" values.
* If there is at least one block with a bucket number greater or equal than zero, then there was a two-level aggregation.
*/
auto max_bucket = bucket_to_blocks.rbegin()->first;
2019-01-21 10:39:24 +00:00
bool has_two_level = max_bucket >= 0;
if (has_two_level)
{
#define M(NAME) \
2018-08-27 17:42:13 +00:00
if (method_chosen == AggregatedDataVariants::Type::NAME) \
method_chosen = AggregatedDataVariants::Type::NAME ## _two_level;
APPLY_FOR_VARIANTS_CONVERTIBLE_TO_TWO_LEVEL(M)
#undef M
}
2012-05-30 01:38:02 +00:00
if (isCancelled())
return;
2015-04-16 14:27:56 +00:00
2017-04-02 17:37:49 +00:00
/// result will destroy the states of aggregate functions in the destructor
result.aggregator = this;
2018-08-27 17:42:13 +00:00
result.init(method_chosen);
result.keys_size = params.keys_size;
result.key_sizes = key_sizes;
2012-05-30 01:38:02 +00:00
bool has_blocks_with_unknown_bucket = bucket_to_blocks.count(-1);
2015-01-17 04:49:13 +00:00
2017-04-02 17:37:49 +00:00
/// First, parallel the merge for the individual buckets. Then we continue merge the data not allocated to the buckets.
if (has_two_level)
{
2017-04-02 17:37:49 +00:00
/** In this case, no_more_keys is not supported due to the fact that
* from different threads it is difficult to update the general state for "other" keys (overflows).
* That is, the keys in the end can be significantly larger than max_rows_to_group_by.
*/
2020-05-23 22:24:01 +00:00
LOG_TRACE(log, "Merging partially aggregated two-level data.");
2012-05-30 01:38:02 +00:00
auto merge_bucket = [&bucket_to_blocks, &result, this](Int32 bucket, Arena * aggregates_pool, ThreadGroupStatusPtr thread_group)
{
if (thread_group)
CurrentThread::attachToIfDetached(thread_group);
2012-05-30 01:38:02 +00:00
for (Block & block : bucket_to_blocks[bucket])
{
if (isCancelled())
return;
2015-04-16 14:27:56 +00:00
#define M(NAME) \
else if (result.type == AggregatedDataVariants::Type::NAME) \
2018-08-27 18:05:28 +00:00
mergeStreamsImpl(block, aggregates_pool, *result.NAME, result.NAME->data.impls[bucket], nullptr, false);
2020-03-08 23:48:08 +00:00
if (false) {} // NOLINT
APPLY_FOR_VARIANTS_TWO_LEVEL(M)
#undef M
else
throw Exception("Unknown aggregated data variant.", ErrorCodes::UNKNOWN_AGGREGATED_DATA_VARIANT);
}
};
std::unique_ptr<ThreadPool> thread_pool;
if (max_threads > 1 && total_input_rows > 100000) /// TODO Make a custom threshold.
thread_pool = std::make_unique<ThreadPool>(max_threads);
for (const auto & bucket_blocks : bucket_to_blocks)
{
const auto bucket = bucket_blocks.first;
if (bucket == -1)
continue;
result.aggregates_pools.push_back(std::make_shared<Arena>());
Arena * aggregates_pool = result.aggregates_pools.back().get();
2020-04-22 18:41:19 +00:00
auto task = [group = CurrentThread::getGroup(), bucket, &merge_bucket, aggregates_pool]{ return merge_bucket(bucket, aggregates_pool, group); };
if (thread_pool)
thread_pool->scheduleOrThrowOnError(task);
else
task();
}
if (thread_pool)
thread_pool->wait();
2020-05-23 22:24:01 +00:00
LOG_TRACE(log, "Merged partially aggregated two-level data.");
}
if (isCancelled())
{
result.invalidate();
return;
}
if (has_blocks_with_unknown_bucket)
{
2020-05-23 22:24:01 +00:00
LOG_TRACE(log, "Merging partially aggregated single-level data.");
bool no_more_keys = false;
BlocksList & blocks = bucket_to_blocks[-1];
for (Block & block : blocks)
{
if (isCancelled())
{
result.invalidate();
return;
}
2015-04-16 14:27:56 +00:00
if (!checkLimits(result.sizeWithoutOverflowRow(), no_more_keys))
break;
if (result.type == AggregatedDataVariants::Type::without_key || block.info.is_overflows)
mergeWithoutKeyStreamsImpl(block, result);
#define M(NAME, IS_TWO_LEVEL) \
else if (result.type == AggregatedDataVariants::Type::NAME) \
2018-08-27 18:05:28 +00:00
mergeStreamsImpl(block, result.aggregates_pool, *result.NAME, result.NAME->data, result.without_key, no_more_keys);
APPLY_FOR_AGGREGATED_VARIANTS(M)
#undef M
else if (result.type != AggregatedDataVariants::Type::without_key)
throw Exception("Unknown aggregated data variant.", ErrorCodes::UNKNOWN_AGGREGATED_DATA_VARIANT);
}
2020-05-23 22:24:01 +00:00
LOG_TRACE(log, "Merged partially aggregated single-level data.");
}
}
Block Aggregator::mergeBlocks(BlocksList & blocks, bool final)
{
if (blocks.empty())
return {};
auto bucket_num = blocks.front().info.bucket_num;
bool is_overflows = blocks.front().info.is_overflows;
2020-05-23 22:24:01 +00:00
LOG_TRACE(log, "Merging partially aggregated blocks (bucket = {}).", bucket_num);
Stopwatch watch;
/** If possible, change 'method' to some_hash64. Otherwise, leave as is.
* Better hash function is needed because during external aggregation,
* we may merge partitions of data with total number of keys far greater than 4 billion.
*/
2018-08-27 17:42:13 +00:00
auto merge_method = method_chosen;
#define APPLY_FOR_VARIANTS_THAT_MAY_USE_BETTER_HASH_FUNCTION(M) \
M(key64) \
M(key_string) \
M(key_fixed_string) \
M(keys128) \
M(keys256) \
M(serialized) \
#define M(NAME) \
if (merge_method == AggregatedDataVariants::Type::NAME) \
merge_method = AggregatedDataVariants::Type::NAME ## _hash64; \
APPLY_FOR_VARIANTS_THAT_MAY_USE_BETTER_HASH_FUNCTION(M)
#undef M
#undef APPLY_FOR_VARIANTS_THAT_MAY_USE_BETTER_HASH_FUNCTION
2017-04-02 17:37:49 +00:00
/// Temporary data for aggregation.
AggregatedDataVariants result;
2017-04-02 17:37:49 +00:00
/// result will destroy the states of aggregate functions in the destructor
result.aggregator = this;
result.init(merge_method);
result.keys_size = params.keys_size;
result.key_sizes = key_sizes;
for (Block & block : blocks)
{
if (isCancelled())
return {};
if (bucket_num >= 0 && block.info.bucket_num != bucket_num)
bucket_num = -1;
if (result.type == AggregatedDataVariants::Type::without_key || is_overflows)
mergeWithoutKeyStreamsImpl(block, result);
#define M(NAME, IS_TWO_LEVEL) \
else if (result.type == AggregatedDataVariants::Type::NAME) \
2018-08-27 18:05:28 +00:00
mergeStreamsImpl(block, result.aggregates_pool, *result.NAME, result.NAME->data, nullptr, false);
APPLY_FOR_AGGREGATED_VARIANTS(M)
#undef M
else if (result.type != AggregatedDataVariants::Type::without_key)
throw Exception("Unknown aggregated data variant.", ErrorCodes::UNKNOWN_AGGREGATED_DATA_VARIANT);
}
if (isCancelled())
return {};
Block block;
if (result.type == AggregatedDataVariants::Type::without_key || is_overflows)
block = prepareBlockAndFillWithoutKey(result, final, is_overflows);
else
block = prepareBlockAndFillSingleLevel(result, final);
/// NOTE: two-level data is not possible here - chooseAggregationMethod chooses only among single-level methods.
if (!final)
{
/// Pass ownership of aggregate function states from result to ColumnAggregateFunction objects in the resulting block.
result.aggregator = nullptr;
}
size_t rows = block.rows();
size_t bytes = block.bytes();
double elapsed_seconds = watch.elapsedSeconds();
2020-05-23 22:24:01 +00:00
LOG_TRACE(log, "Merged partially aggregated blocks. {} rows, {}. in {} sec. ({} rows/sec., {}/sec.)",
2020-05-30 21:35:52 +00:00
rows, ReadableSize(bytes),
2020-05-23 21:26:45 +00:00
elapsed_seconds, rows / elapsed_seconds,
2020-05-30 21:35:52 +00:00
ReadableSize(bytes / elapsed_seconds));
if (isCancelled())
return {};
block.info.bucket_num = bucket_num;
return block;
}
template <typename Method>
void NO_INLINE Aggregator::convertBlockToTwoLevelImpl(
Method & method,
Arena * pool,
ColumnRawPtrs & key_columns,
const Block & source,
std::vector<Block> & destinations) const
{
2019-01-21 10:39:24 +00:00
typename Method::State state(key_columns, key_sizes, aggregation_state_cache);
size_t rows = source.rows();
size_t columns = source.columns();
/// Create a 'selector' that will contain bucket index for every row. It will be used to scatter rows to buckets.
IColumn::Selector selector(rows);
/// For every row.
for (size_t i = 0; i < rows; ++i)
{
if constexpr (Method::low_cardinality_optimization)
{
if (state.isNullAt(i))
{
selector[i] = 0;
continue;
}
}
2019-01-21 10:39:24 +00:00
/// Calculate bucket number from row hash.
auto hash = state.getHash(method.data, i, *pool);
auto bucket = method.data.getBucketFromHash(hash);
selector[i] = bucket;
}
size_t num_buckets = destinations.size();
for (size_t column_idx = 0; column_idx < columns; ++column_idx)
{
const ColumnWithTypeAndName & src_col = source.getByPosition(column_idx);
MutableColumns scattered_columns = src_col.column->scatter(num_buckets, selector);
for (size_t bucket = 0, size = num_buckets; bucket < size; ++bucket)
{
if (!scattered_columns[bucket]->empty())
{
Block & dst = destinations[bucket];
dst.info.bucket_num = bucket;
dst.insert({std::move(scattered_columns[bucket]), src_col.type, src_col.name});
}
/** Inserted columns of type ColumnAggregateFunction will own states of aggregate functions
* by holding shared_ptr to source column. See ColumnAggregateFunction.h
*/
}
}
}
std::vector<Block> Aggregator::convertBlockToTwoLevel(const Block & block)
{
if (!block)
return {};
AggregatedDataVariants data;
ColumnRawPtrs key_columns(params.keys_size);
2017-04-02 17:37:49 +00:00
/// Remember the columns we will work with
for (size_t i = 0; i < params.keys_size; ++i)
key_columns[i] = block.safeGetByPosition(i).column.get();
2018-08-27 17:42:13 +00:00
AggregatedDataVariants::Type type = method_chosen;
data.keys_size = params.keys_size;
data.key_sizes = key_sizes;
#define M(NAME) \
else if (type == AggregatedDataVariants::Type::NAME) \
type = AggregatedDataVariants::Type::NAME ## _two_level;
2020-03-08 23:48:08 +00:00
if (false) {} // NOLINT
APPLY_FOR_VARIANTS_CONVERTIBLE_TO_TWO_LEVEL(M)
#undef M
else
throw Exception("Unknown aggregated data variant.", ErrorCodes::UNKNOWN_AGGREGATED_DATA_VARIANT);
data.init(type);
size_t num_buckets = 0;
#define M(NAME) \
else if (data.type == AggregatedDataVariants::Type::NAME) \
num_buckets = data.NAME->data.NUM_BUCKETS;
2020-03-08 23:48:08 +00:00
if (false) {} // NOLINT
APPLY_FOR_VARIANTS_TWO_LEVEL(M)
#undef M
else
throw Exception("Unknown aggregated data variant.", ErrorCodes::UNKNOWN_AGGREGATED_DATA_VARIANT);
std::vector<Block> splitted_blocks(num_buckets);
#define M(NAME) \
else if (data.type == AggregatedDataVariants::Type::NAME) \
convertBlockToTwoLevelImpl(*data.NAME, data.aggregates_pool, \
2019-08-11 21:45:18 +00:00
key_columns, block, splitted_blocks);
2020-03-08 23:48:08 +00:00
if (false) {} // NOLINT
APPLY_FOR_VARIANTS_TWO_LEVEL(M)
#undef M
else
throw Exception("Unknown aggregated data variant.", ErrorCodes::UNKNOWN_AGGREGATED_DATA_VARIANT);
return splitted_blocks;
}
2015-12-06 19:42:28 +00:00
template <typename Method, typename Table>
2017-12-01 20:21:35 +00:00
void NO_INLINE Aggregator::destroyImpl(Table & table) const
{
table.forEachMapped([&](AggregateDataPtr & data)
{
2017-04-02 17:37:49 +00:00
/** If an exception (usually a lack of memory, the MemoryTracker throws) arose
* after inserting the key into a hash table, but before creating all states of aggregate functions,
* then data will be equal nullptr.
*/
if (nullptr == data)
return;
for (size_t i = 0; i < params.aggregates_size; ++i)
2020-06-07 22:46:58 +00:00
aggregate_functions[i]->destroy(data + offsets_of_aggregate_states[i]);
data = nullptr;
});
2015-12-06 19:42:28 +00:00
}
void Aggregator::destroyWithoutKey(AggregatedDataVariants & result) const
{
AggregatedDataWithoutKey & res_data = result.without_key;
2015-12-06 19:42:28 +00:00
if (nullptr != res_data)
{
for (size_t i = 0; i < params.aggregates_size; ++i)
2020-06-07 22:46:58 +00:00
aggregate_functions[i]->destroy(res_data + offsets_of_aggregate_states[i]);
2015-12-06 19:42:28 +00:00
res_data = nullptr;
}
2012-05-30 01:38:02 +00:00
}
void Aggregator::destroyAllAggregateStates(AggregatedDataVariants & result)
{
2020-03-08 21:40:00 +00:00
if (result.empty())
return;
2020-05-23 22:24:01 +00:00
LOG_TRACE(log, "Destroying aggregate states");
2017-04-02 17:37:49 +00:00
/// In what data structure is the data aggregated?
if (result.type == AggregatedDataVariants::Type::without_key || params.overflow_row)
destroyWithoutKey(result);
#define M(NAME, IS_TWO_LEVEL) \
else if (result.type == AggregatedDataVariants::Type::NAME) \
2017-12-01 20:21:35 +00:00
destroyImpl<decltype(result.NAME)::element_type>(result.NAME->data);
2020-03-08 23:48:08 +00:00
if (false) {} // NOLINT
APPLY_FOR_AGGREGATED_VARIANTS(M)
#undef M
else if (result.type != AggregatedDataVariants::Type::without_key)
throw Exception("Unknown aggregated data variant.", ErrorCodes::UNKNOWN_AGGREGATED_DATA_VARIANT);
}
2015-04-16 14:27:56 +00:00
void Aggregator::setCancellationHook(const CancellationHook cancellation_hook)
{
isCancelled = cancellation_hook;
2015-04-16 14:27:56 +00:00
}
2011-09-19 01:42:16 +00:00
}