Aggregation Pipeline

The Aggregation Pipeline is MongoDB's framework for transforming and combining documents through a sequence of stages, conceptually similar to a SQL SELECT with GROUP BY, JOIN, and window functions. A pipeline is an array of stage operators applied in order; the output of one stage feeds the next.

Common stages

• $match: filter documents (equivalent to WHERE)
• $project: reshape and select fields (equivalent to SELECT)
• $group: group and aggregate (equivalent to GROUP BY)
• $sort: sort documents (equivalent to ORDER BY)
• $limit, $skip: pagination
• $lookup: left outer join to another collection
• $unwind: explode array fields into one document per element
• $facet: run multiple sub-pipelines in parallel on the same input
• $graphLookup: recursive graph traversal
• $merge, $out: write results to another collection

Performance considerations

• Place $match and $project early to reduce documents and fields flowing downstream.
• Use indexes to support early $match and $sort stages.
• $lookup is expensive; pre-denormalising data sometimes beats join-style queries.
• Use explain() to inspect the executed plan and see which stages used indexes.