Time-series collections in MongoDB

Introduction

Time-series collections help store data that changes over time, like temperature or stock prices, in an easy and efficient way.

Tracking temperature readings every minute from sensors.

Recording stock prices every second during trading hours.

Logging website visitor counts every hour.

Monitoring machine performance data continuously.

Syntax

MongoDB

db.createCollection("collectionName", {
  timeseries: {
    timeField: "timestamp",
    metaField: "metadata",
    granularity: "seconds" // or "minutes", "hours"
  }
})

timeField is the field that stores the time of each record.

metaField is optional and stores extra info about the data.

Examples

This creates a time-series collection for temperature data recorded every minute with sensor ID as metadata.

MongoDB

db.createCollection("temperatureReadings", {
  timeseries: {
    timeField: "time",
    metaField: "sensorId",
    granularity: "minutes"
  }
})

This creates a time-series collection for stock prices recorded every second without extra metadata.

MongoDB

db.createCollection("stockPrices", {
  timeseries: {
    timeField: "tradeTime",
    granularity: "seconds"
  }
})

Sample Program

This example creates a time-series collection for machine metrics, inserts three records, and then queries data for machine M1 sorted by time.

MongoDB

db.createCollection("machineMetrics", {
  timeseries: {
    timeField: "recordedAt",
    metaField: "machineId",
    granularity: "seconds"
  }
})

// Insert sample data
db.machineMetrics.insertMany([
  { recordedAt: new Date("2024-06-01T10:00:00Z"), machineId: "M1", temperature: 75, rpm: 1500 },
  { recordedAt: new Date("2024-06-01T10:00:05Z"), machineId: "M1", temperature: 76, rpm: 1520 },
  { recordedAt: new Date("2024-06-01T10:00:10Z"), machineId: "M2", temperature: 70, rpm: 1400 }
])

// Query data
db.machineMetrics.find({ machineId: "M1" }).sort({ recordedAt: 1 })

Important Notes

Time-series collections optimize storage and queries for data with timestamps.

Choosing the right granularity helps MongoDB manage data efficiently.

Use metaField to store information that stays the same for many records, like sensor or machine ID.

Summary

Time-series collections store data that changes over time efficiently.

They require a time field and optionally a metadata field.

They help with fast queries and better storage for time-based data.

Practice

1. What is the main purpose of a time-series collection in MongoDB?

easy

A. To manage user authentication and roles

B. To store large binary files like images and videos

C. To efficiently store and query data that changes over time

D. To create backups of the database automatically

Solution

1. Step 1: Understand time-series data

   Time-series data is data recorded over time, like temperature or stock prices.

2. Step 2: Identify MongoDB time-series collection use

   MongoDB time-series collections optimize storage and queries for such time-based data.

3. Final Answer:

   To efficiently store and query data that changes over time -> Option C

4. Quick Check:

   Time-series collections = efficient time-based data storage [OK]

Hint: Time-series collections are for data changing over time [OK]

Common Mistakes:

• Confusing time-series collections with file storage
• Thinking time-series collections manage users
• Assuming they handle backups automatically

2. Which of the following is required when creating a time-series collection in MongoDB?

easy

A. A field that stores binary data

B. A unique index on the _id field

C. A field that stores user roles

D. A field that stores the timestamp of each record

Solution

1. Step 1: Recall time-series collection requirements

   MongoDB requires a time field to track when each data point was recorded.

2. Step 2: Check other options

   Unique _id index exists by default but is not specific to time-series; user roles and binary data fields are unrelated.

3. Final Answer:

   A field that stores the timestamp of each record -> Option D

4. Quick Check:

   Time field required = A field that stores the timestamp of each record [OK]

Hint: Time field is mandatory for time-series collections [OK]

Common Mistakes:

• Choosing unrelated fields like user roles
• Confusing default indexes with required time field
• Thinking binary data field is required

3. Given the following command, what will be the result?

db.createCollection("sensorData", {
  timeseries: {
    timeField: "timestamp",
    metaField: "deviceId"
  }
})

medium

A. Creates a time-series collection with timestamp and deviceId fields

B. Throws a syntax error due to missing options

C. Creates a regular collection named sensorData

D. Creates a capped collection instead of time-series

Solution

1. Step 1: Analyze the createCollection command

   The command uses the timeseries option with timeField and metaField specified.

2. Step 2: Understand MongoDB behavior

   This creates a time-series collection named sensorData with timestamp as time and deviceId as metadata.

3. Final Answer:

   Creates a time-series collection with timestamp and deviceId fields -> Option A

4. Quick Check:

   timeseries option creates time-series collection [OK]

Hint: timeseries option creates time-series collection [OK]

Common Mistakes:

• Thinking it creates a regular or capped collection
• Assuming syntax error due to missing options
• Confusing metaField with required timeField

4. You tried to create a time-series collection but got an error. Which of the following is the most likely cause?

db.createCollection("temps", {
  timeseries: {
    metaField: "location"
  }
})

medium

A. Missing required timeField in timeseries options

B. metaField cannot be used in time-series collections

C. Collection name is invalid

D. timeseries option is misspelled

Solution

1. Step 1: Check timeseries options in the command

   The command specifies metaField but omits the required timeField.

2. Step 2: Recall MongoDB requirements

   timeField is mandatory for time-series collections; missing it causes an error.

3. Final Answer:

   Missing required timeField in timeseries options -> Option A

4. Quick Check:

   timeField missing = error [OK]

Hint: Always include timeField when creating time-series collections [OK]

Common Mistakes:

• Thinking metaField is not allowed
• Assuming collection name is invalid
• Believing timeseries option is misspelled

5. You have a time-series collection with fields timestamp (timeField) and sensorId (metaField). You want to query the average temperature for sensorId 'S1' over the last 24 hours. Which aggregation pipeline is correct?

db.sensorData.aggregate([
  { $match: { sensorId: 'S1', timestamp: { $gte: new Date(Date.now() - 86400000) } } },
  { $group: { _id: null, avgTemp: { $avg: "$temperature" } } }
])

hard

A. Groups by sensorId instead of null

B. Correctly filters by sensorId and timestamp, then averages temperature

C. Uses $sort instead of $match for filtering

D. Filters only by timestamp, ignoring sensorId

Solution

1. Step 1: Analyze the $match stage

   The pipeline filters documents where sensorId is 'S1' and timestamp is within last 24 hours.

2. Step 2: Analyze the $group stage

   It groups all matched documents (_id: null) and calculates average temperature.

3. Final Answer:

   Correctly filters by sensorId and timestamp, then averages temperature -> Option B

4. Quick Check:

   Filter by sensorId and time, then average = Correctly filters by sensorId and timestamp, then averages temperature [OK]

Hint: Match filters, group aggregates for averages [OK]

Common Mistakes:

• Forgetting to filter by sensorId
• Using $sort instead of $match for filtering
• Grouping by sensorId when only one sensor is queried