How to create a performance system that count tags across a large dynamic dataset in MongoDB?

To create a performance system that counts tags across a large dynamic dataset in MongoDB, use indexes on array fields combined with aggregation pipelines and the explain() method to monitor query performance.

Syntax

// Create index on array field
db.collection.createIndex({"arrayField": 1});

// Count tags with aggregation
db.collection.aggregate([
    { $unwind: "$arrayField" },
    { $group: { _id: "$arrayField", count: { $sum: 1 } } }
]);

// Monitor performance
db.collection.find(query).explain("executionStats");

Sample Data

Let us create a collection with documents containing subject tags −

db.demo278.createIndex({"Subjects": 1});

db.demo278.insertMany([
    {"Subjects": ["MySQL", "MongoDB", "Java"]},
    {"Subjects": ["C", "C++"]},
    {"Subjects": ["Spring", "Hibernate"]}
]);

{
    "createdCollectionAutomatically": true,
    "numIndexesBefore": 1,
    "numIndexesAfter": 2,
    "ok": 1
}

Display Sample Data

db.demo278.find();

{ "_id": ObjectId("5e49096edd099650a5401a55"), "Subjects": ["MySQL", "MongoDB", "Java"] }
{ "_id": ObjectId("5e490978dd099650a5401a56"), "Subjects": ["C", "C++"] }
{ "_id": ObjectId("5e490984dd099650a5401a57"), "Subjects": ["Spring", "Hibernate"] }

Method 1: Count Specific Tags with Performance Analysis

Query documents containing specific tags and analyze performance −

db.demo278.find({Subjects: {$in: ["Spring", "MongoDB"]}}).explain("executionStats");

{
    "queryPlanner": {
        "plannerVersion": 1,
        "namespace": "test.demo278",
        "indexFilterSet": false,
        "parsedQuery": {
            "Subjects": {
                "$in": ["MongoDB", "Spring"]
            }
        },
        "winningPlan": {
            "stage": "FETCH",
            "inputStage": {
                "stage": "IXSCAN",
                "keyPattern": {"Subjects": 1},
                "indexName": "Subjects_1",
                "isMultiKey": true,
                "indexBounds": {
                    "Subjects": [
                        "["MongoDB", "MongoDB"]",
                        "["Spring", "Spring"]"
                    ]
                }
            }
        }
    },
    "ok": 1
}

Method 2: Count All Tags Using Aggregation

Create a comprehensive tag counting system −

db.demo278.aggregate([
    { $unwind: "$Subjects" },
    { $group: { _id: "$Subjects", count: { $sum: 1 } } },
    { $sort: { count: -1 } }
]);

{ "_id": "C", "count": 1 }
{ "_id": "C++", "count": 1 }
{ "_id": "Hibernate", "count": 1 }
{ "_id": "Java", "count": 1 }
{ "_id": "MongoDB", "count": 1 }
{ "_id": "MySQL", "count": 1 }
{ "_id": "Spring", "count": 1 }

Key Performance Points

• The IXSCAN stage indicates the query uses the index efficiently
• isMultiKey: true confirms the index supports array fields
• Use explain("executionStats") for detailed execution metrics
• Aggregation pipelines with $unwind are ideal for tag counting

Conclusion

Create indexes on array fields and use aggregation pipelines for efficient tag counting. Monitor performance with explain() to ensure queries use indexes and scale effectively across large datasets.