Radom data is useful to test many scenarios and check our code stability in many ways. We can get random documents from given collection.

Solution 1: Using $sample

Randomly selects the specified number of documents from the input documents. The $sample stage has the following syntax:

{ $sample: { size:  } }

N is the number of documents to randomly select.

From 3.2+ MongoDB, you can get N random docs from a collection using the $sample aggregation pipeline operator:

db.mycoll.aggregate([{ $sample: { size: N } }])

N can be any number.

If you want to select the random document(s) from a filtered subset of the collection, prepend a $match stage to the pipeline:

// Get one random document matching {a: 10} from the mycoll collection.
db.mycoll.aggregate([
    { $match: { a: 10 } },
    { $sample: { size: N } }
])

Note: $sample may get the same document more than once

Solution 2: $rand

$rand returns a random float between 0 and 1. $rand has the following syntax:

{ $rand: {} }

The $rand operator doesn’t take any arguments.

Using query

db.mycoll.find( {$expr: { $lt: [0.5, {$rand: {} }] }})

Using Aggregation (4.4.2+)

db.mycoll.aggregate(
   [   
      { $match: { $expr: { $lt: [0.5, {$rand: {} } ] } } }     
   ]
)

The threshold of 0.5 in the less than ($lt) comparison means that $expr will be true for about half the documents.

The number of documents selected is different each time. If you need to select an exact number of documents, consider using $sample instead of $rand.

Source: Stackoverflow