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
122
123
124
125
126
127
128
129
|
/**
* Copyright (C) 2018-present MongoDB, Inc.
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the Server Side Public License, version 1,
* as published by MongoDB, Inc.
*
* 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
* Server Side Public License for more details.
*
* You should have received a copy of the Server Side Public License
* along with this program. If not, see
* <http://www.mongodb.com/licensing/server-side-public-license>.
*
* 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 Server Side 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 <vector>
#include "mongo/bson/bsonobj.h"
#include "mongo/db/namespace_string.h"
#include "mongo/s/catalog/type_chunk.h"
#include "mongo/s/catalog/type_tags.h"
#include "mongo/s/shard_id.h"
#include "mongo/s/shard_key_pattern.h"
namespace mongo {
class InitialSplitPolicy {
public:
/**
* For new collections which use hashed shard keys, we can can pre-split the range of possible
* hashes into a large number of chunks, and distribute them evenly at creation time.
*
* Until we design a better initialization scheme, the most performant way to pre-split is to
* make one big chunk for each shard and migrate them one at a time. Because of this:
* - 'initialSplitPoints' is populated with the split points to use on the primary shard to
* produce the initial "big chunks."
* - 'finalSplitPoints' is populated with the additional split points to use on the "big chunks"
* after the "big chunks" have been spread evenly across shards through migrations.
*/
static void calculateHashedSplitPointsForEmptyCollection(
const ShardKeyPattern& shardKeyPattern,
bool isEmpty,
int numShards,
int numInitialChunks,
std::vector<BSONObj>* initialSplitPoints,
std::vector<BSONObj>* finalSplitPoints);
struct ShardCollectionConfig {
std::vector<ChunkType> chunks;
const auto& collVersion() const {
return chunks.back().getVersion();
}
};
/**
* Produces the initial chunks that need to be written for a collection which is being
* newly-sharded. The function performs some basic validation of the input parameters, but there
* is no checking whether the collection contains any data or not.
*
* Chunks are assigned to a shard in a round-robin fashion, numContiguousChunksPerShard (k)
* chunks at a time. For example, the first k chunks are assigned to the first available shard,
* and the next k chunks are assigned to the second available shard and so on.
* numContiguousChunksPerShard should only be > 1 when we do not pre-split the range
* into larger chunks and then split the resulting chunks on the destination shards as in
* configSvrShardCollection, thus should be equal the number of final split points + 1 divided
* by the number of initial split points + 1. It serves to preserve the ordering/contigousness
* of chunks when split by shardSvrShardCollection so that its yields the exact same shard
* assignments as configSvrShardCollection.
*/
static ShardCollectionConfig generateShardCollectionInitialChunks(
const NamespaceString& nss,
const ShardKeyPattern& shardKeyPattern,
const ShardId& databasePrimaryShardId,
const Timestamp& validAfter,
const std::vector<BSONObj>& splitPoints,
const std::vector<ShardId>& allShardIds,
const int numContiguousChunksPerShard = 1);
/**
* Produces the initial chunks that need to be written for a collection which is being
* newly-sharded based on the given tags. Chunks that do not correspond to any pre-defined
* zones are assigned to available shards in a round-robin fashion.
*/
static ShardCollectionConfig generateShardCollectionInitialZonedChunks(
const NamespaceString& nss,
const ShardKeyPattern& shardKeyPattern,
const Timestamp& validAfter,
const std::vector<TagsType>& tags,
const StringMap<std::vector<ShardId>>& tagToShards,
const std::vector<ShardId>& allShardIds);
/**
* Creates the first chunks for a newly sharded collection.
* Returns the created chunks.
*/
static ShardCollectionConfig createFirstChunks(OperationContext* opCtx,
const NamespaceString& nss,
const ShardKeyPattern& shardKeyPattern,
const ShardId& primaryShardId,
const std::vector<BSONObj>& splitPoints,
const std::vector<TagsType>& tags,
const bool distributeInitialChunks,
const int numContiguousChunksPerShard = 1);
/**
* Writes to the config server the first chunks for a newly sharded collection.
*/
static void writeFirstChunksToConfig(
OperationContext* opCtx, const InitialSplitPolicy::ShardCollectionConfig& initialChunks);
};
} // namespace mongo
|
