# Proposed changes Issue Number: close #6238 Co-authored-by: HappenLee <happenlee@hotmail.com> Co-authored-by: stdpain <34912776+stdpain@users.noreply.github.com> Co-authored-by: Zhengguo Yang <yangzhgg@gmail.com> Co-authored-by: wangbo <506340561@qq.com> Co-authored-by: emmymiao87 <522274284@qq.com> Co-authored-by: Pxl <952130278@qq.com> Co-authored-by: zhangstar333 <87313068+zhangstar333@users.noreply.github.com> Co-authored-by: thinker <zchw100@qq.com> Co-authored-by: Zeno Yang <1521564989@qq.com> Co-authored-by: Wang Shuo <wangshuo128@gmail.com> Co-authored-by: zhoubintao <35688959+zbtzbtzbt@users.noreply.github.com> Co-authored-by: Gabriel <gabrielleebuaa@gmail.com> Co-authored-by: xinghuayu007 <1450306854@qq.com> Co-authored-by: weizuo93 <weizuo@apache.org> Co-authored-by: yiguolei <guoleiyi@tencent.com> Co-authored-by: anneji-dev <85534151+anneji-dev@users.noreply.github.com> Co-authored-by: awakeljw <993007281@qq.com> Co-authored-by: taberylyang <95272637+taberylyang@users.noreply.github.com> Co-authored-by: Cui Kaifeng <48012748+azurenake@users.noreply.github.com> ## Problem Summary: ### 1. Some code from clickhouse **ClickHouse is an excellent implementation of the vectorized execution engine database, so here we have referenced and learned a lot from its excellent implementation in terms of data structure and function implementation. We are based on ClickHouse v19.16.2.2 and would like to thank the ClickHouse community and developers.** The following comment has been added to the code from Clickhouse, eg: // This file is copied from // https://github.com/ClickHouse/ClickHouse/blob/master/src/Interpreters/AggregationCommon.h // and modified by Doris ### 2. Support exec node and query: * vaggregation_node * vanalytic_eval_node * vassert_num_rows_node * vblocking_join_node * vcross_join_node * vempty_set_node * ves_http_scan_node * vexcept_node * vexchange_node * vintersect_node * vmysql_scan_node * vodbc_scan_node * volap_scan_node * vrepeat_node * vschema_scan_node * vselect_node * vset_operation_node * vsort_node * vunion_node * vhash_join_node You can run exec engine of SSB/TPCH and 70% TPCDS stand query test set. ### 3. Data Model Vec Exec Engine Support **Dup/Agg/Unq** table, Support Block Reader Vectorized. Segment Vec is working in process. ### 4. How to use 1. Set the environment variable `set enable_vectorized_engine = true; `(required) 2. Set the environment variable `set batch_size = 4096; ` (recommended) ### 5. Some diff from origin exec engine https://github.com/doris-vectorized/doris-vectorized/issues/294 ## Checklist(Required) 1. Does it affect the original behavior: (No) 2. Has unit tests been added: (Yes) 3. Has document been added or modified: (No) 4. Does it need to update dependencies: (No) 5. Are there any changes that cannot be rolled back: (Yes)
198 lines
7.1 KiB
C++
198 lines
7.1 KiB
C++
// Licensed to the Apache Software Foundation (ASF) under one
|
|
// or more contributor license agreements. See the NOTICE file
|
|
// distributed with this work for additional information
|
|
// regarding copyright ownership. The ASF licenses this file
|
|
// to you under the Apache License, Version 2.0 (the
|
|
// "License"); you may not use this file except in compliance
|
|
// with the License. You may obtain a copy of the License at
|
|
//
|
|
// http://www.apache.org/licenses/LICENSE-2.0
|
|
//
|
|
// Unless required by applicable law or agreed to in writing,
|
|
// software distributed under the License is distributed on an
|
|
// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
|
|
// KIND, either express or implied. See the License for the
|
|
// specific language governing permissions and limitations
|
|
// under the License.
|
|
|
|
#include "exec/blocking_join_node.h"
|
|
|
|
#include <sstream>
|
|
|
|
#include "exprs/expr.h"
|
|
#include "gen_cpp/PlanNodes_types.h"
|
|
#include "runtime/row_batch.h"
|
|
#include "runtime/runtime_state.h"
|
|
#include "util/runtime_profile.h"
|
|
|
|
namespace doris {
|
|
|
|
BlockingJoinNode::BlockingJoinNode(const std::string& node_name, const TJoinOp::type join_op,
|
|
ObjectPool* pool, const TPlanNode& tnode,
|
|
const DescriptorTbl& descs)
|
|
: ExecNode(pool, tnode, descs), _node_name(node_name), _join_op(join_op),
|
|
_left_side_eos(false) {}
|
|
|
|
Status BlockingJoinNode::init(const TPlanNode& tnode, RuntimeState* state) {
|
|
return ExecNode::init(tnode, state);
|
|
}
|
|
|
|
BlockingJoinNode::~BlockingJoinNode() {
|
|
// _left_batch must be cleaned up in close() to ensure proper resource freeing.
|
|
DCHECK(_left_batch == nullptr);
|
|
}
|
|
|
|
Status BlockingJoinNode::prepare(RuntimeState* state) {
|
|
SCOPED_TIMER(_runtime_profile->total_time_counter());
|
|
RETURN_IF_ERROR(ExecNode::prepare(state));
|
|
|
|
_build_pool.reset(new MemPool(mem_tracker().get()));
|
|
_build_timer = ADD_TIMER(runtime_profile(), "BuildTime");
|
|
_left_child_timer = ADD_TIMER(runtime_profile(), "LeftChildTime");
|
|
_build_row_counter = ADD_COUNTER(runtime_profile(), "BuildRows", TUnit::UNIT);
|
|
_left_child_row_counter = ADD_COUNTER(runtime_profile(), "LeftChildRows", TUnit::UNIT);
|
|
|
|
_result_tuple_row_size = _row_descriptor.tuple_descriptors().size() * sizeof(Tuple*);
|
|
|
|
// pre-compute the tuple index of build tuples in the output row
|
|
int num_left_tuples = child(0)->row_desc().tuple_descriptors().size();
|
|
int num_build_tuples = child(1)->row_desc().tuple_descriptors().size();
|
|
|
|
_build_tuple_size = num_build_tuples;
|
|
_build_tuple_idx.reserve(_build_tuple_size);
|
|
|
|
for (int i = 0; i < _build_tuple_size; ++i) {
|
|
TupleDescriptor* build_tuple_desc = child(1)->row_desc().tuple_descriptors()[i];
|
|
_build_tuple_idx.push_back(_row_descriptor.get_tuple_idx(build_tuple_desc->id()));
|
|
}
|
|
|
|
_probe_tuple_row_size = num_left_tuples * sizeof(Tuple*);
|
|
_build_tuple_row_size = num_build_tuples * sizeof(Tuple*);
|
|
|
|
_left_batch.reset(new RowBatch(child(0)->row_desc(), state->batch_size(), mem_tracker().get()));
|
|
return Status::OK();
|
|
}
|
|
|
|
Status BlockingJoinNode::close(RuntimeState* state) {
|
|
// TODO(zhaochun): avoid double close
|
|
// if (is_closed()) return Status::OK();
|
|
_left_batch.reset();
|
|
ExecNode::close(state);
|
|
return Status::OK();
|
|
}
|
|
|
|
void BlockingJoinNode::build_side_thread(RuntimeState* state, std::promise<Status>* status) {
|
|
status->set_value(construct_build_side(state));
|
|
}
|
|
|
|
Status BlockingJoinNode::open(RuntimeState* state) {
|
|
RETURN_IF_ERROR(ExecNode::open(state));
|
|
SCOPED_TIMER(_runtime_profile->total_time_counter());
|
|
// RETURN_IF_ERROR(Expr::open(_conjuncts, state));
|
|
|
|
RETURN_IF_CANCELLED(state);
|
|
// TODO(zhaochun)
|
|
// RETURN_IF_ERROR(state->check_query_state());
|
|
|
|
_eos = false;
|
|
|
|
// Kick-off the construction of the build-side table in a separate
|
|
// thread, so that the left child can do any initialisation in parallel.
|
|
// Only do this if we can get a thread token. Otherwise, do this in the
|
|
// main thread
|
|
std::promise<Status> build_side_status;
|
|
|
|
add_runtime_exec_option("Join Build-Side Prepared Asynchronously");
|
|
std::thread(bind(&BlockingJoinNode::build_side_thread, this, state, &build_side_status))
|
|
.detach();
|
|
|
|
// Open the left child so that it may perform any initialisation in parallel.
|
|
// Don't exit even if we see an error, we still need to wait for the build thread
|
|
// to finish.
|
|
Status open_status = child(0)->open(state);
|
|
|
|
// Blocks until ConstructBuildSide has returned, after which the build side structures
|
|
// are fully constructed.
|
|
RETURN_IF_ERROR(build_side_status.get_future().get());
|
|
// We can close the right child to release its resources because its input has been
|
|
// fully consumed.
|
|
child(1)->close(state);
|
|
|
|
RETURN_IF_ERROR(open_status);
|
|
|
|
// Seed left child in preparation for get_next().
|
|
while (true) {
|
|
RETURN_IF_ERROR(child(0)->get_next(state, _left_batch.get(), &_left_side_eos));
|
|
COUNTER_UPDATE(_left_child_row_counter, _left_batch->num_rows());
|
|
_left_batch_pos = 0;
|
|
|
|
if (_left_batch->num_rows() == 0) {
|
|
if (_left_side_eos) {
|
|
init_get_next(nullptr /* eos */);
|
|
_eos = true;
|
|
break;
|
|
}
|
|
|
|
_left_batch->reset();
|
|
continue;
|
|
} else {
|
|
_current_left_child_row = _left_batch->get_row(_left_batch_pos++);
|
|
init_get_next(_current_left_child_row);
|
|
break;
|
|
}
|
|
}
|
|
|
|
return Status::OK();
|
|
}
|
|
|
|
void BlockingJoinNode::debug_string(int indentation_level, std::stringstream* out) const {
|
|
*out << std::string(indentation_level * 2, ' ');
|
|
*out << _node_name;
|
|
*out << "(eos=" << (_eos ? "true" : "false") << " left_batch_pos=" << _left_batch_pos;
|
|
add_to_debug_string(indentation_level, out);
|
|
ExecNode::debug_string(indentation_level, out);
|
|
*out << ")";
|
|
}
|
|
|
|
std::string BlockingJoinNode::get_left_child_row_string(TupleRow* row) {
|
|
std::stringstream out;
|
|
out << "[";
|
|
int* _build_tuple_idx_ptr = &_build_tuple_idx[0];
|
|
|
|
for (int i = 0; i < row_desc().tuple_descriptors().size(); ++i) {
|
|
if (i != 0) {
|
|
out << " ";
|
|
}
|
|
|
|
int* is_build_tuple =
|
|
std::find(_build_tuple_idx_ptr, _build_tuple_idx_ptr + _build_tuple_size, i);
|
|
|
|
if (is_build_tuple != _build_tuple_idx_ptr + _build_tuple_size) {
|
|
out << Tuple::to_string(nullptr, *row_desc().tuple_descriptors()[i]);
|
|
} else {
|
|
out << Tuple::to_string(row->get_tuple(i), *row_desc().tuple_descriptors()[i]);
|
|
}
|
|
}
|
|
|
|
out << "]";
|
|
return out.str();
|
|
}
|
|
|
|
// This function is replaced by codegen
|
|
void BlockingJoinNode::create_output_row(TupleRow* out, TupleRow* left, TupleRow* build) {
|
|
uint8_t* out_ptr = reinterpret_cast<uint8_t*>(out);
|
|
if (left == nullptr) {
|
|
memset(out_ptr, 0, _probe_tuple_row_size);
|
|
} else {
|
|
memcpy(out_ptr, left, _probe_tuple_row_size);
|
|
}
|
|
|
|
if (build == nullptr) {
|
|
memset(out_ptr + _probe_tuple_row_size, 0, _build_tuple_row_size);
|
|
} else {
|
|
memcpy(out_ptr + _probe_tuple_row_size, build, _build_tuple_row_size);
|
|
}
|
|
}
|
|
|
|
} // namespace doris
|