Computed pattern for pre-aggregation in MongoDB - Step-by-Step Execution

Concept Flow - Computed pattern for pre-aggregation

Start: Raw Data Collection

↓

Define Computed Fields

↓

Apply Aggregation Pipeline

↓

Compute and Store Aggregated Results

↓

Use Pre-Aggregated Data for Queries

↓

End

This flow shows how raw data is processed by defining computed fields, aggregating them, storing results, and then using these pre-aggregated results for faster queries.

Execution Sample

MongoDB

db.sales.aggregate([
  { $project: {
      item: 1,
      totalPrice: { $multiply: ["$price", "$quantity"] }
    }
  },
  { $group: { _id: "$item", totalSales: { $sum: "$totalPrice" } } }
])

This MongoDB aggregation computes total price per sale and then sums total sales per item.

Execution Table

Step	Pipeline Stage	Input Document	Computed Field	Group Key	Aggregated Value
1	$project	{"item":"apple", "price":2, "quantity":3}	totalPrice = 2*3=6
2	$project	{"item":"banana", "price":1, "quantity":5}	totalPrice = 1*5=5
3	$project	{"item":"apple", "price":2, "quantity":2}	totalPrice = 2*2=4
4	$group	{"item":"apple", "totalPrice":6}		apple	6
5	$group	{"item":"banana", "totalPrice":5}		banana	5
6	$group	{"item":"apple", "totalPrice":4}		apple	6+4=10
7	End				Final aggregated totals: apple=10, banana=5

💡 All documents processed; aggregation complete with pre-aggregated totals per item.

Variable Tracker

Variable	Start	After 1	After 2	After 3	After 4	After 5	After 6	Final
totalPrice	N/A	6	5	4	N/A	N/A	N/A	N/A
totalSales_apple	0	6	6	10	10	10	10	10
totalSales_banana	0	0	5	5	5	5	5	5

Key Moments - 2 Insights

Why do we compute 'totalPrice' before grouping?

Computing 'totalPrice' in the $project stage creates a new field for each document, which can then be summed in the $group stage. This separation ensures accurate aggregation as shown in execution_table rows 1-3 for computation and 4-6 for grouping.

How does the $group stage accumulate totals?

The $group stage sums 'totalPrice' values for each unique 'item' key. For example, in rows 4 and 6, 'apple' totals are added step-by-step to reach the final sum 10.

Visual Quiz - 3 Questions

Look at the execution_table, what is the computed 'totalPrice' for the second document at step 2?

A 6

B 4

C 5

D 10

💡 Hint

Check the 'Computed Field' column at step 2 in the execution_table.

At which step does the total sales for 'apple' become 10?

A Step 5

B Step 6

C Step 4

D Step 7

💡 Hint

Look at the 'Aggregated Value' for 'apple' in the execution_table rows.

If the quantity of the first 'apple' document changed from 3 to 4, what would be the new 'totalPrice' at step 1?

A 8

B 10

C 6

D 12

💡 Hint

Multiply price 2 by new quantity 4 as shown in the 'Computed Field' logic.

Concept Snapshot

MongoDB pre-aggregation uses $project to compute new fields (e.g., totalPrice = price * quantity).
Then $group sums these computed fields by keys (e.g., item).
This pattern speeds up queries by storing aggregated results.
Use $project before $group to prepare data for aggregation.
Final output is grouped totals ready for fast retrieval.

Full Transcript

This visual execution traces the MongoDB computed pattern for pre-aggregation. Starting with raw sales data, each document computes a totalPrice by multiplying price and quantity in the $project stage. Then, the $group stage sums totalPrice values grouped by item names. The execution table shows each step: computing totalPrice for each document, then accumulating sums per item. Variable tracking reveals how totalPrice and totalSales for each item change after each step. Key moments clarify why computation happens before grouping and how sums accumulate. The visual quiz tests understanding of computed values and aggregation steps. This pattern helps store pre-aggregated data for faster queries.