be/src/exec/spill/spill_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 | | |
20 | | #include <atomic> |
21 | | #include <memory> |
22 | | #include <string> |
23 | | |
24 | | #include "core/block/block.h" |
25 | | #include "io/fs/file_writer.h" |
26 | | #include "runtime/runtime_profile.h" |
27 | | #include "runtime/workload_management/resource_context.h" |
28 | | namespace doris { |
29 | | #include "common/compile_check_begin.h" |
30 | | class RuntimeState; |
31 | | |
32 | | class SpillDataDir; |
33 | | class SpillWriter { |
34 | | public: |
35 | | SpillWriter(std::shared_ptr<ResourceContext> resource_context, RuntimeProfile* profile, |
36 | | int64_t id, size_t batch_size, SpillDataDir* data_dir, const std::string& dir) |
37 | 2.04k | : data_dir_(data_dir), |
38 | 2.04k | stream_id_(id), |
39 | 2.04k | batch_size_(batch_size), |
40 | 2.04k | _resource_ctx(std::move(resource_context)) { |
41 | | // Directory path format specified in SpillStreamManager::register_spill_stream: |
42 | | // storage_root/spill/query_id/partitioned_hash_join-node_id-task_id-stream_id/0 |
43 | 2.04k | file_path_ = dir + "/0"; |
44 | 2.04k | RuntimeProfile* common_profile = profile->get_child("CommonCounters"); |
45 | 2.04k | DCHECK(common_profile != nullptr); |
46 | 2.04k | _memory_used_counter = common_profile->get_counter("MemoryUsage"); |
47 | 2.04k | } |
48 | | |
49 | | Status open(); |
50 | | |
51 | | Status close(); |
52 | | |
53 | | Status write(RuntimeState* state, const Block& block, size_t& written_bytes); |
54 | | |
55 | 0 | int64_t get_id() const { return stream_id_; } |
56 | | |
57 | 6.09k | int64_t get_written_bytes() const { return total_written_bytes_; } |
58 | | |
59 | 2.06k | const std::string& get_file_path() const { return file_path_; } |
60 | | |
61 | 2.04k | void set_counters(RuntimeProfile* operator_profile) { |
62 | 2.04k | RuntimeProfile* custom_profile = operator_profile->get_child("CustomCounters"); |
63 | 2.04k | _write_file_timer = custom_profile->get_counter("SpillWriteFileTime"); |
64 | 2.04k | _serialize_timer = custom_profile->get_counter("SpillWriteSerializeBlockTime"); |
65 | 2.04k | _write_block_counter = custom_profile->get_counter("SpillWriteBlockCount"); |
66 | 2.04k | _write_block_bytes_counter = custom_profile->get_counter("SpillWriteBlockBytes"); |
67 | 2.04k | _write_file_total_size = custom_profile->get_counter("SpillWriteFileBytes"); |
68 | 2.04k | _write_file_current_size = custom_profile->get_counter("SpillWriteFileCurrentBytes"); |
69 | 2.04k | _write_rows_counter = custom_profile->get_counter("SpillWriteRows"); |
70 | 2.04k | } |
71 | | |
72 | | private: |
73 | | Status _write_internal(const Block& block, size_t& written_bytes); |
74 | | |
75 | | // not owned, point to the data dir of this rowset |
76 | | // for checking disk capacity when write data to disk. |
77 | | SpillDataDir* data_dir_ = nullptr; |
78 | | std::atomic_bool closed_ = false; |
79 | | int64_t stream_id_; |
80 | | size_t batch_size_; |
81 | | size_t max_sub_block_size_ = 0; |
82 | | std::string file_path_; |
83 | | std::unique_ptr<doris::io::FileWriter> file_writer_; |
84 | | |
85 | | size_t written_blocks_ = 0; |
86 | | int64_t total_written_bytes_ = 0; |
87 | | std::string meta_; |
88 | | |
89 | | RuntimeProfile::Counter* _write_file_timer = nullptr; |
90 | | RuntimeProfile::Counter* _serialize_timer = nullptr; |
91 | | RuntimeProfile::Counter* _write_block_counter = nullptr; |
92 | | RuntimeProfile::Counter* _write_block_bytes_counter = nullptr; |
93 | | RuntimeProfile::Counter* _write_file_total_size = nullptr; |
94 | | RuntimeProfile::Counter* _write_file_current_size = nullptr; |
95 | | RuntimeProfile::Counter* _write_rows_counter = nullptr; |
96 | | RuntimeProfile::Counter* _memory_used_counter = nullptr; |
97 | | |
98 | | std::shared_ptr<ResourceContext> _resource_ctx = nullptr; |
99 | | }; |
100 | | using SpillWriterUPtr = std::unique_ptr<SpillWriter>; |
101 | | } // namespace doris |
102 | | |
103 | | #include "common/compile_check_end.h" |