be/src/exec/sink/writer/async_result_writer.h
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 | | #pragma once |
19 | | #include <concurrentqueue.h> |
20 | | |
21 | | #include <condition_variable> |
22 | | #include <queue> // IWYU pragma: keep |
23 | | |
24 | | #include "exec/sink/writer/async_writer_queue_admission.h" |
25 | | #include "exec/sink/writer/result_writer.h" |
26 | | #include "exprs/vexpr_fwd.h" |
27 | | #include "runtime/memory/thread_mem_tracker_mgr.h" |
28 | | #include "runtime/runtime_profile.h" |
29 | | |
30 | | namespace doris { |
31 | | class ObjectPool; |
32 | | class RowDescriptor; |
33 | | class RuntimeState; |
34 | | class TDataSink; |
35 | | class TExpr; |
36 | | |
37 | | class Dependency; |
38 | | class PipelineTask; |
39 | | |
40 | | class Block; |
41 | | |
42 | | /* |
43 | | * In the pipeline execution engine, there are usually a large number of io operations on the sink side that |
44 | | * will block the limited execution threads of the pipeline execution engine, resulting in a sharp performance |
45 | | * degradation of the pipeline execution engine when there are import tasks. |
46 | | * |
47 | | * So all ResultWriter in Sink should use AsyncResultWriter to do the real IO task in thread pool to keep the |
48 | | * pipeline execution engine performance. |
49 | | * |
50 | | * The Sub class of AsyncResultWriter need to impl two virtual function |
51 | | * * Status open() the first time IO work like: create file/ connect network |
52 | | * * Status write() do the real IO work for block |
53 | | */ |
54 | | class AsyncResultWriter : public ResultWriter { |
55 | | public: |
56 | | AsyncResultWriter(const VExprContextSPtrs& output_expr_ctxs, std::shared_ptr<Dependency> dep, |
57 | | std::shared_ptr<Dependency> fin_dep); |
58 | | |
59 | | void force_close(Status s); |
60 | | |
61 | 0 | Status init(RuntimeState* state) override { return Status::OK(); } |
62 | | |
63 | | virtual Status open(RuntimeState* state, RuntimeProfile* operator_profile) = 0; |
64 | | |
65 | | // sink the block data to data queue, it is async |
66 | | Status sink(Block* block, bool eos); |
67 | | |
68 | | // Add the IO thread task process block() to thread pool to dispose the IO |
69 | | Status start_writer(RuntimeState* state, RuntimeProfile* operator_profile); |
70 | | |
71 | 1 | Status get_writer_status() { return _writer_status.status(); } |
72 | | |
73 | | void set_low_memory_mode(); |
74 | | |
75 | 0 | void wait_for_processing_before_next_sink() { |
76 | 0 | _queue_admission.wait_for_processing_before_next_sink(); |
77 | 0 | } |
78 | | |
79 | | protected: |
80 | | Status _projection_block(Block& input_block, Block* output_block); |
81 | | const VExprContextSPtrs& _vec_output_expr_ctxs; |
82 | | RuntimeProfile* _operator_profile = nullptr; // not owned, set when open |
83 | | |
84 | | std::unique_ptr<Block> _get_free_block(Block*, size_t rows); |
85 | | |
86 | | private: |
87 | | struct QueuedBlock { |
88 | | std::unique_ptr<Block> block; |
89 | | ReservedMemoryToken reservation; |
90 | | bool eos = false; |
91 | | }; |
92 | | |
93 | | void process_block(RuntimeState* state, RuntimeProfile* operator_profile); |
94 | 5 | [[nodiscard]] bool _data_queue_is_available() const { |
95 | 5 | return _queue_admission.is_available(_data_queue.size()); |
96 | 5 | } |
97 | 4 | [[nodiscard]] bool _is_finished() const { return !_writer_status.ok() || _eos; } |
98 | | void _set_ready_to_finish(); |
99 | | |
100 | | void _return_free_block(std::unique_ptr<Block>); |
101 | | QueuedBlock _get_block_from_queue(); |
102 | | void _notify_block_processed(); |
103 | | |
104 | | std::mutex _m; |
105 | | std::condition_variable _cv; |
106 | | std::deque<QueuedBlock> _data_queue; |
107 | | // Default value is ok |
108 | | AtomicStatus _writer_status; |
109 | | bool _eos = false; |
110 | | AsyncWriterQueueAdmission _queue_admission; |
111 | | std::atomic_bool _low_memory_mode = false; |
112 | | |
113 | | std::shared_ptr<Dependency> _dependency; |
114 | | std::shared_ptr<Dependency> _finish_dependency; |
115 | | |
116 | | moodycamel::ConcurrentQueue<std::unique_ptr<Block>> _free_blocks; |
117 | | RuntimeProfile::Counter* _memory_used_counter = nullptr; |
118 | | }; |
119 | | |
120 | | } // namespace doris |