1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
|
/**
* Copyright (C) 2018 MongoDB Inc.
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License, version 3,
* as published by the Free Software Foundation.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU Affero General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*
* As a special exception, the copyright holders give permission to link the
* code of portions of this program with the OpenSSL library under certain
* conditions as described in each individual source file and distribute
* linked combinations including the program with the OpenSSL library. You
* must comply with the GNU Affero General Public License in all respects
* for all of the code used other than as permitted herein. If you modify
* file(s) with this exception, you may extend this exception to your
* version of the file(s), but you are not obligated to do so. If you do not
* wish to do so, delete this exception statement from your version. If you
* delete this exception statement from all source files in the program,
* then also delete it in the license file.
*/
#pragma once
#include "mongo/db/pipeline/document_source_exchange_gen.h"
#include "mongo/db/pipeline/lite_parsed_pipeline.h"
#include "mongo/db/pipeline/pipeline.h"
#include "mongo/db/pipeline/pipeline.h"
#include "mongo/s/catalog/type_chunk.h"
#include "mongo/s/query/cluster_client_cursor_impl.h"
#include "mongo/s/shard_id.h"
namespace mongo {
namespace cluster_aggregation_planner {
/**
* Represents the two halves of a pipeline that will execute in a sharded cluster. 'shardsPipeline'
* will execute in parallel on each shard, and 'mergePipeline' will execute on the merge host -
* either one of the shards or a mongos.
*/
struct SplitPipeline {
SplitPipeline(std::unique_ptr<Pipeline, PipelineDeleter> shardsPipeline,
std::unique_ptr<Pipeline, PipelineDeleter> mergePipeline,
boost::optional<BSONObj> shardCursorsSortSpec)
: shardsPipeline(std::move(shardsPipeline)),
mergePipeline(std::move(mergePipeline)),
shardCursorsSortSpec(std::move(shardCursorsSortSpec)) {}
std::unique_ptr<Pipeline, PipelineDeleter> shardsPipeline;
std::unique_ptr<Pipeline, PipelineDeleter> mergePipeline;
// If set, the cursors from the shards are expected to be sorted according to this spec, and to
// have populated a "$sortKey" metadata field which can be used to compare the results.
boost::optional<BSONObj> shardCursorsSortSpec;
};
/**
* Split the current Pipeline into a Pipeline for each shard, and a Pipeline that combines the
* results within a merging process. This call also performs optimizations with the aim of reducing
* computing time and network traffic when a pipeline has been split into two pieces.
*
* The 'mergePipeline' returned as part of the SplitPipeline here is not ready to execute until the
* 'shardsPipeline' has been sent to the shards and cursors have been established. Once cursors have
* been established, the merge pipeline can be made executable by calling 'addMergeCursorsSource()'
*/
SplitPipeline splitPipeline(std::unique_ptr<Pipeline, PipelineDeleter> pipeline);
/**
* Creates a new DocumentSourceMergeCursors from the provided 'remoteCursors' and adds it to the
* front of 'mergePipeline'.
*/
void addMergeCursorsSource(Pipeline* mergePipeline,
const LiteParsedPipeline&,
BSONObj cmdSentToShards,
std::vector<RemoteCursor> remoteCursors,
const std::vector<ShardId>& targetedShards,
boost::optional<BSONObj> shardCursorsSortSpec,
executor::TaskExecutor*);
/**
* Builds a ClusterClientCursor which will execute 'pipeline'. If 'pipeline' consists entirely of
* $skip and $limit stages, the pipeline is eliminated entirely and replaced with a RouterExecStage
* tree that does same thing but will avoid using a RouterStagePipeline. Avoiding a
* RouterStagePipeline will remove an expensive conversion from BSONObj -> Document for each result.
*/
ClusterClientCursorGuard buildClusterCursor(OperationContext* opCtx,
std::unique_ptr<Pipeline, PipelineDeleter> pipeline,
ClusterClientCursorParams&&);
struct ShardDistributionInfo {
// If we want to send data to the shards which would own the data, 'logicalShardKeyAtSplitPoint'
// describes which of the fields to use to determine what the final shard key will be. For
// example, if the merging pipeline renames "x" to "out_shard_key" and then uses $out to output
// to a collection sharded by {out_shard_key: 1}, 'logicalShardKeyAtSplitPoint' will be {x: 1}.
ShardKeyPattern logicalShardKeyAtSplitPoint;
// This map describes which shard is going to receive which range. The keys are the shard ids.
StringMap<std::vector<ChunkRange>> partitions;
};
struct ShardedExchangePolicy {
ExchangePolicyEnum policy;
// Only set if the policy is ranged.
boost::optional<ShardDistributionInfo> shardDistributionInfo;
};
/**
* If the merging pipeline is eligible for an $exchange merge optimization, returns the information
* required to set that up.
*/
boost::optional<ShardedExchangePolicy> checkIfEligibleForExchange(OperationContext* opCtx,
const Pipeline* mergePipeline);
} // namespace cluster_aggregation_planner
} // namespace mongo
|