be/src/exec/pipeline/pipeline.cpp
Line | Count | Source |
1 | | // Licensed to the Apache Software Foundation (ASF) under one |
2 | | // or more contributor license agreements. See the NOTICE file |
3 | | // distributed with this work for additional information |
4 | | // regarding copyright ownership. The ASF licenses this file |
5 | | // to you under the Apache License, Version 2.0 (the |
6 | | // "License"); you may not use this file except in compliance |
7 | | // with the License. You may obtain a copy of the License at |
8 | | // |
9 | | // http://www.apache.org/licenses/LICENSE-2.0 |
10 | | // |
11 | | // Unless required by applicable law or agreed to in writing, |
12 | | // software distributed under the License is distributed on an |
13 | | // "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY |
14 | | // KIND, either express or implied. See the License for the |
15 | | // specific language governing permissions and limitations |
16 | | // under the License. |
17 | | |
18 | | #include "exec/pipeline/pipeline.h" |
19 | | |
20 | | #include <memory> |
21 | | #include <string> |
22 | | #include <utility> |
23 | | |
24 | | #include "exec/exchange/local_exchange_source_operator.h" |
25 | | #include "exec/operator/operator.h" |
26 | | #include "exec/pipeline/pipeline_fragment_context.h" |
27 | | #include "exec/pipeline/pipeline_task.h" |
28 | | |
29 | | namespace doris { |
30 | | |
31 | 144k | void Pipeline::_init_profile() { |
32 | 144k | auto s = fmt::format("Pipeline (pipeline id={})", _pipeline_id); |
33 | 144k | _pipeline_profile = std::make_unique<RuntimeProfile>(std::move(s)); |
34 | 144k | } |
35 | | |
36 | | bool Pipeline::need_to_local_exchange(const DataDistribution target_data_distribution, |
37 | 6 | const int idx) const { |
38 | 6 | if (!target_data_distribution.need_local_exchange()) { |
39 | 2 | return false; |
40 | 2 | } |
41 | | // If serial operator exists after `idx`-th operator, we should not improve parallelism. |
42 | 4 | if (std::any_of(_operators.begin() + idx, _operators.end(), |
43 | 4 | [&](OperatorPtr op) -> bool { return op->is_serial_operator(); })) { |
44 | 0 | return false; |
45 | 0 | } |
46 | | // If all operators are serial and sink is not serial, we should improve parallelism for sink. |
47 | 4 | if (std::all_of(_operators.begin(), _operators.end(), |
48 | 4 | [&](OperatorPtr op) -> bool { return op->is_serial_operator(); })) { |
49 | 1 | if (!_sink->is_serial_operator()) { |
50 | 0 | return true; |
51 | 0 | } |
52 | 3 | } else if (std::any_of(_operators.begin(), _operators.end(), |
53 | 4 | [&](OperatorPtr op) -> bool { return op->is_serial_operator(); })) { |
54 | | // If non-serial operators exist, we should improve parallelism for those. |
55 | 0 | return true; |
56 | 0 | } |
57 | 4 | if (auto local_exchange_source = |
58 | 4 | std::dynamic_pointer_cast<LocalExchangeSourceOperatorX>(_operators.front()); |
59 | 4 | local_exchange_source && is_hash_exchange(target_data_distribution.distribution_type)) { |
60 | 1 | const auto source_exchange_type = local_exchange_source->exchange_type(); |
61 | 1 | if (source_exchange_type != ExchangeType::NOOP && !is_hash_exchange(source_exchange_type)) { |
62 | 1 | return true; |
63 | 1 | } |
64 | 1 | } |
65 | 3 | if (target_data_distribution.distribution_type != ExchangeType::BUCKET_HASH_SHUFFLE && |
66 | 3 | target_data_distribution.distribution_type != ExchangeType::HASH_SHUFFLE) { |
67 | | // Always do local exchange if non-hash-partition exchanger is required. |
68 | | // For example, `PASSTHROUGH` exchanger is always required to distribute data evenly. |
69 | 1 | return true; |
70 | 2 | } else if (_operators.front()->is_serial_operator()) { |
71 | 0 | DCHECK(std::all_of(_operators.begin(), _operators.end(), |
72 | 0 | [&](OperatorPtr op) -> bool { return op->is_serial_operator(); }) && |
73 | 0 | _sink->is_serial_operator()) |
74 | 0 | << debug_string(); |
75 | | // All operators and sink are serial in this path. |
76 | 0 | return false; |
77 | 2 | } else { |
78 | 2 | return _data_distribution.distribution_type != target_data_distribution.distribution_type && |
79 | 2 | !(is_hash_exchange(_data_distribution.distribution_type) && |
80 | 0 | is_hash_exchange(target_data_distribution.distribution_type)); |
81 | 2 | } |
82 | 3 | } |
83 | | |
84 | 144k | Status Pipeline::add_operator(OperatorPtr& op, const int parallelism) { |
85 | 144k | if (parallelism > 0 && op->is_serial_operator()) { |
86 | 0 | set_num_tasks(parallelism); |
87 | 0 | } |
88 | 144k | op->set_parallel_tasks(num_tasks()); |
89 | 144k | _operators.emplace_back(op); |
90 | 144k | if (op->is_source()) { |
91 | 72.1k | std::reverse(_operators.begin(), _operators.end()); |
92 | 72.1k | } |
93 | 144k | return Status::OK(); |
94 | 144k | } |
95 | | |
96 | 3 | Status Pipeline::prepare(RuntimeState* state) { |
97 | 3 | RETURN_IF_ERROR(_operators.back()->prepare(state)); |
98 | 3 | RETURN_IF_ERROR(_sink->prepare(state)); |
99 | 3 | _name.append(std::to_string(id())); |
100 | 3 | _name.push_back('-'); |
101 | 4 | for (auto& op : _operators) { |
102 | 4 | _name.append(std::to_string(op->node_id())); |
103 | 4 | _name.append(op->get_name()); |
104 | 4 | } |
105 | 3 | _name.push_back('-'); |
106 | 3 | _name.append(std::to_string(_sink->node_id())); |
107 | 3 | _name.append(_sink->get_name()); |
108 | 3 | return Status::OK(); |
109 | 3 | } |
110 | | |
111 | 144k | Status Pipeline::set_sink(DataSinkOperatorPtr& sink) { |
112 | 144k | if (_sink) { |
113 | 0 | return Status::InternalError("set sink twice"); |
114 | 0 | } |
115 | 144k | if (!sink->is_sink()) { |
116 | 0 | return Status::InternalError("should set a sink operator but {}", typeid(sink).name()); |
117 | 0 | } |
118 | 144k | _sink = sink; |
119 | 144k | return Status::OK(); |
120 | 144k | } |
121 | | |
122 | 0 | void Pipeline::make_all_runnable(PipelineId wake_by) { |
123 | 0 | DBUG_EXECUTE_IF("Pipeline::make_all_runnable.sleep", { |
124 | 0 | auto pipeline_id = DebugPoints::instance()->get_debug_param_or_default<int32_t>( |
125 | 0 | "Pipeline::make_all_runnable.sleep", "pipeline_id", -1); |
126 | 0 | if (pipeline_id == id()) { |
127 | 0 | LOG(WARNING) << "Pipeline::make_all_runnable.sleep sleep 10s"; |
128 | 0 | sleep(10); |
129 | 0 | } |
130 | 0 | }); |
131 | |
|
132 | 0 | if (_sink->count_down_destination()) { |
133 | 0 | for (auto* task : _tasks) { |
134 | 0 | if (task) { |
135 | 0 | task->set_wake_up_early(wake_by); |
136 | 0 | task->unblock_all_dependencies(); |
137 | 0 | } |
138 | 0 | } |
139 | 0 | } |
140 | 0 | } |
141 | | |
142 | | } // namespace doris |