MongoDB $collStats Stage

Example 1 — Sample Data

Seed five orders and one secondary index so storage and count stats have something meaningful to report.

db.orders.drop()
db.orders.createIndex({ customerId: 1 })
db.orders.insertMany([
  { customerId: "C001", product: "Keyboard", amount: 45,  status: "shipped" },
  { customerId: "C002", product: "Mouse",    amount: 25,  status: "pending" },
  { customerId: "C001", product: "Monitor",  amount: 220, status: "shipped" },
  { customerId: "C003", product: "Webcam",   amount: 60,  status: "pending" },
  { customerId: "C002", product: "Headset",  amount: 80,  status: "shipped" }
])

How the Data Is Stored in MongoDB

After the insert, MongoDB keeps five documents in orders. Run find() to see them:

db.orders.find()

[
  { customerId: "C001", product: "Keyboard", amount: 45,  status: "shipped" },
  { customerId: "C002", product: "Mouse",    amount: 25,  status: "pending" },
  { customerId: "C001", product: "Monitor",  amount: 220, status: "shipped" },
  { customerId: "C003", product: "Webcam",   amount: 60,  status: "pending" },
  { customerId: "C002", product: "Headset",  amount: 80,  status: "shipped" }
]
// plus indexes: _id_ and customerId_1

Read the important fields like this:

How It Works

1. Reset the collection. drop() starts clean for repeatable demos.
2. Add an index. customerId gives storageStats a second index to measure.
3. Insert five orders. Small sample—enough for count and size fields to be non-zero.

Example 2 — storageStats

Goal: get collection size, document count, and index totals, then keep only the fields a dashboard needs.

db.orders.aggregate([
  { $collStats: { storageStats: {} } },
  {
    $project: {
      ns: 1,
      host: 1,
      count: "$storageStats.count",
      dataSize: "$storageStats.size",
      avgDoc: "$storageStats.avgObjectSize",
      indexes: "$storageStats.nindexes",
      totalIndexSize: "$storageStats.totalIndexSize"
    }
  }
])

{
  ns: "test.orders",
  host: "localhost:27017",
  count: 5,
  dataSize: 1234,        // bytes (scale 1)
  avgDoc: 246,
  indexes: 2,            // _id + customerId
  totalIndexSize: 8192
}

How It Works

1. First stage only. $collStats must lead the pipeline.
2. Enable storage. storageStats: {} fills size and index fields (bytes by default).
3. Trim with $project. Keep the metrics you care about; drop WiredTiger noise.
4. Note: Exact byte values vary by engine and version; the field names stay the same.

Example 3 — Count + Scale

Goal: get a fast metadata count, and report storage sizes in kilobytes with scale: 1024.

// Count only — lightweight
db.orders.aggregate([
  { $collStats: { count: {} } }
])
// → { ns, host, localTime, count: 5 }

// Storage in kilobytes + count together
db.orders.aggregate([
  {
    $collStats: {
      storageStats: { scale: 1024 },
      count: {}
    }
  },
  {
    $project: {
      count: 1,
      dataSizeKB: "$storageStats.size",
      indexSizeKB: "$storageStats.totalIndexSize",
      scaleFactor: "$storageStats.scaleFactor"
    }
  }
])

// count only
{ ns: "test.orders", count: 5, … }

// scaled storage
{
  count: 5,
  dataSizeKB: 1.2,       // illustrative
  indexSizeKB: 8,
  scaleFactor: 1024
}

// Reminder:
// scale: 1    → bytes (default)
// scale: 1024 → kilobytes

How It Works

1. Count alone. count: {} is a quick metadata total—not a full collection scan of each doc.
2. Scale sizes. scale: 1024 divides reported sizes so dashboards show KB.
3. Combine options. One $collStats object can enable storage and count together.
4. Exact filtered totals? Prefer $count after a $match.