Aggregation Pipeline and Sharded Collections聚合管道和分片集合

Behavior行为

If the pipeline starts with an exact $match on a shard key, and the pipeline does not contain $out or $lookup stages, the entire pipeline runs on the matching shard only.如果管道从分片键上的精确$match开始，并且管道不包含$out或$lookup阶段，则整个管道仅在匹配的分片上运行。

When aggregation operations run on multiple shards, the results are routed to the mongos to be merged, except in the following cases:当聚合操作在多个分片上运行时，结果会被路由到要合并的mongos，以下情况除外：

• If the pipeline includes the $out stage, the merge runs on the shard where the output collection lives.如果管道包含$out阶段，则合并将在输出集合所在的分片上运行。
• If the pipeline includes the $lookup stage that references an unsharded collection, the merge runs on the shard where the unsharded collection lives.如果管道包含引用未分片集合的$lookup阶段，则合并将在未分片集合所在的分片上运行。
• If the pipeline includes a sorting or grouping stage, and the allowDiskUse setting is enabled, the merge runs on a randomly-selected shard.如果管道包括排序或分组阶段，并且启用了allowDiskUse设置，则合并将在随机选择的分片上运行。

Optimization优化

When splitting the aggregation pipeline into two parts, the pipeline is split to ensure that the shards perform as many stages as possible with consideration for optimization.当将聚合管道拆分为两部分时，会对管道进行拆分，以确保分片在考虑优化的情况下执行尽可能多的阶段。

To see how the pipeline was split, include the explain option in the db.collection.aggregate() method.要查看管道是如何拆分的，请在db.collection.aggregate()方法中包含explain选项。

Optimizations are subject to change between releases.优化可能会在不同版本之间发生变化。