Docs Home → Develop Applications → MongoDB Manual
Map-Reduce and Sharded Collections
Note
Aggregation Pipeline as an Alternative to Map-Reduce
Starting in MongoDB 5.0, map-reduce is deprecated:
Instead of map-reduce, you should use an aggregation pipeline. Aggregation pipelines provide better performance and usability than map-reduce.
You can rewrite map-reduce operations using aggregation pipeline stages, such as
$group,$merge, and others.For map-reduce operations that require custom functionality, you can use the
$accumulatorand$functionaggregation operators. You can use those operators to define custom aggregation expressions in JavaScript.
For examples of aggregation pipeline alternatives to map-reduce, see:
Map-reduce supports operations on sharded collections, both as an input
and as an output. This section describes the behaviors of
mapReduce specific to sharded collections.
Sharded Collection as Input
When using sharded collection as the input for a map-reduce operation,
mongos will automatically dispatch the map-reduce job to
each shard in parallel. There is no special option
required. mongos will wait for jobs on all shards to
finish.
Sharded Collection as Output
If the out field for mapReduce has the sharded
value, MongoDB shards the output collection using the _id field as
the shard key.
To output to a sharded collection:
If the output collection does not exist, create the sharded collection first.
Starting in version 4.0, if the output collection already exists but is not sharded, map-reduce fails.
For a new or an empty sharded collection, MongoDB uses the results of the first stage of the map-reduce operation to create the initial chunks distributed among the shards.
mongosdispatches, in parallel, a map-reduce post-processing job to every shard that owns a chunk. During the post-processing, each shard will pull the results for its own chunks from the other shards, run the final reduce/finalize, and write locally to the output collection.
Note
During later map-reduce jobs, MongoDB splits chunks as needed.
Balancing of chunks for the output collection is automatically prevented during post-processing to avoid concurrency issues.