Atlas data federation concept in MongoDB - Step-by-Step Execution

Concept Flow - Atlas data federation concept

User Query Sent

↓

Atlas Data Federation Receives Query

↓

Identify Data Sources

↓

Query Each Source

↓

Aggregate Results

↓

Return Combined Data to User

The flow shows how a user query is sent to Atlas Data Federation, which identifies data sources, queries them, aggregates results, and returns combined data.

Execution Sample

MongoDB

db.users.aggregate([
  { $match: { status: "active" } },
  { $project: { name: 1, email: 1 } }
])

This query filters active users and projects only their name and email from federated data sources.

Execution Table

Step	Action	Data Source	Query Part	Result
1	Receive user query	N/A	Match status: active	Query accepted
2	Identify data sources	Source A, Source B	N/A	Sources identified
3	Send query to Source A	Source A	Match status: active	3 matching documents
4	Send query to Source B	Source B	Match status: active	2 matching documents
5	Aggregate results	N/A	Combine documents	5 documents combined
6	Project fields	N/A	Select name, email	5 documents with name and email
7	Return results to user	N/A	N/A	5 documents returned

💡 All sources queried and results combined; query completed successfully.

Variable Tracker

Variable	Start	After Step 3	After Step 4	After Step 5	Final
results	[]	[3 docs from Source A]	[3 docs + 2 docs]	[5 combined docs]	[5 docs with projected fields]

Key Moments - 3 Insights

How does Atlas Data Federation know which data sources to query?

It identifies relevant data sources based on the query and configured federated collections, as shown in Step 2 of the execution_table.

Are queries run on all data sources at once or one by one?

Queries are sent to each data source individually but can happen in parallel; Steps 3 and 4 show querying Source A and Source B separately.

How are results combined from different sources?

Results from each source are aggregated into a single combined set before projection and return, as shown in Step 5.

Visual Quiz - 3 Questions

Look at the execution_table, what happens at Step 4?

A User query received

B Results aggregated from all sources

C Query sent to Source B to match active status

D Fields projected to name and email

💡 Hint

Check the 'Action' and 'Data Source' columns at Step 4 in execution_table.

At which step are the results combined from all sources?

A Step 5

B Step 3

C Step 7

D Step 2

💡 Hint

Look for 'Aggregate results' action in the execution_table.

If Source B had no matching documents, how would the 'results' variable change after Step 4?

A It would contain documents from Source B only

B It would contain only documents from Source A

C It would be empty

D It would contain documents from both sources

💡 Hint

Refer to variable_tracker and consider what happens if Source B returns zero documents.

Concept Snapshot

Atlas Data Federation lets you query multiple data sources as one.
You send a query to Atlas, it finds sources, queries each, and combines results.
You get unified data without moving it.
Useful for querying across MongoDB Atlas, S3, and more.
Supports filtering, projection, and aggregation across sources.

Full Transcript

Atlas Data Federation allows users to run a single query that accesses multiple data sources. The process starts when the user sends a query. Atlas Data Federation receives it and identifies which data sources are relevant. It then sends the query to each source separately. Each source returns matching documents. These results are aggregated into one combined set. Finally, the combined data is projected to include only requested fields and returned to the user. This lets users work with data from different places as if it were one database, without moving or copying data.