Why Numeric field types in Elasticsearch? - Purpose & Use Cases

The Big Idea

What if your price searches could be instant and always correct, without extra work?

The Scenario

Imagine you have a huge list of product prices stored as plain text in your search database. You want to find all products cheaper than $50, but since prices are text, you have to manually convert and compare each one every time.

The Problem

This manual way is slow because every search needs extra steps to convert text to numbers. It's also error-prone since text sorting treats '100' as less than '50'. This causes wrong search results and wastes time.

The Solution

Using numeric field types in Elasticsearch means prices are stored as real numbers. Elasticsearch can quickly compare, sort, and filter these values without extra work. This makes searches fast, accurate, and easy.

Before vs After

✗ Before

GET /products/_search
{
  "query": {
    "range": {
      "price": {
        "lt": "50"
      }
    }
  }
}

✓ After

PUT /products
{
  "mappings": {
    "properties": {
      "price": { "type": "float" }
    }
  }
}

GET /products/_search
{
  "query": {
    "range": {
      "price": {
        "lt": 50
      }
    }
  }
}

What It Enables

It enables lightning-fast and precise numeric searches, filters, and sorting on your data.

Real Life Example

Online stores use numeric field types to quickly find products under a certain price or within a price range, giving shoppers instant results.

Key Takeaways

Storing numbers as text causes slow and wrong searches.

Numeric field types store real numbers for fast, accurate queries.

This improves search speed and result quality in Elasticsearch.

Practice

1.

Which numeric field type in Elasticsearch should you use to store whole numbers without decimals?

easy

A. double

B. integer

C. float

D. text

Solution

1. Step 1: Understand numeric field types

   Integer type stores whole numbers without decimals, suitable for counts or IDs.

2. Step 2: Compare with other types

   Float and double store decimal numbers; text is for strings, not numbers.

3. Final Answer:

   integer -> Option B

4. Quick Check:

   Whole numbers = integer [OK]

Hint: Whole numbers use integer type, decimals need float/double [OK]

Common Mistakes:

• Choosing float or double for whole numbers
• Using text type for numbers
• Confusing integer with long for small numbers

2.

Which of the following is the correct way to define a long numeric field in an Elasticsearch mapping?

{
  "mappings": {
    "properties": {
      "user_id": {
        "type": "_____"
      }
    }
  }
}

easy

A. "long"

B. "integer"

C. "float"

D. "keyword"

Solution

1. Step 1: Identify the field type for large whole numbers

   Long type stores larger whole numbers than integer, suitable for big IDs.

2. Step 2: Match the correct type string in mapping

   Use "long" as the type value in the mapping JSON for such fields.

3. Final Answer:

   "long" -> Option A

4. Quick Check:

   Big whole numbers = "long" [OK]

Hint: Use "long" for large whole numbers in mapping [OK]

Common Mistakes:

• Using "integer" for very large numbers
• Using "keyword" which is for text
• Using "float" for whole numbers

3.

Given this Elasticsearch mapping snippet, what type of number can price store?

{
  "mappings": {
    "properties": {
      "price": {
        "type": "float"
      }
    }
  }
}

medium

A. Boolean true/false values

B. Only whole numbers

C. Text values representing numbers

D. Decimal numbers with floating point

Solution

1. Step 1: Check the type assigned to price

   The type is "float", which stores decimal numbers with floating point precision.

2. Step 2: Understand what float means

   Float allows decimals, unlike integer or long which store whole numbers only.

3. Final Answer:

   Decimal numbers with floating point -> Option D

4. Quick Check:

   Float type = decimals allowed [OK]

Hint: Float type stores decimal numbers, not just whole numbers [OK]

Common Mistakes:

• Thinking float stores only whole numbers
• Confusing float with text or boolean
• Assuming float stores exact decimal precision always

4.

What is wrong with this Elasticsearch mapping for a numeric field?

{
  "mappings": {
    "properties": {
      "score": {
        "type": "integer",
        "index": "not_analyzed"
      }
    }
  }
}

medium

A. 'type' should be 'float' instead of 'integer'

B. 'score' field name is not allowed

C. 'index' parameter is invalid for numeric fields

D. Mapping syntax is missing a closing brace

Solution

1. Step 1: Review the 'index' parameter usage

   Numeric fields do not support 'index' with 'not_analyzed'; this is for text fields.

2. Step 2: Confirm correct mapping for numeric fields

   Numeric fields are indexed by default and do not use 'not_analyzed'.

3. Final Answer:

   'index' parameter is invalid for numeric fields -> Option C

4. Quick Check:

   Numeric fields don't use 'index': 'not_analyzed' [OK]

Hint: Don't use 'index':'not_analyzed' on numeric fields [OK]

Common Mistakes:

• Using text field options on numeric fields
• Changing type unnecessarily
• Ignoring Elasticsearch mapping syntax rules

5.

You want to store product weights that can be very large and may have decimals. Which Elasticsearch numeric field type is best?

hard

A. double

B. long

C. float

D. integer

Solution

1. Step 1: Analyze the requirements for product weights

   Weights can be very large and have decimals, so the type must support large range and decimals.

2. Step 2: Compare float and double types

   Float supports decimals but has less precision and smaller range than double.

3. Step 3: Choose the best type for large decimal numbers

   Double supports large decimal numbers with higher precision, suitable for weights.

4. Final Answer:

   double -> Option A

5. Quick Check:

   Large decimals = double type [OK]

Hint: Use double for large decimal numbers needing precision [OK]

Common Mistakes:

• Choosing integer or long which don't support decimals
• Using float which may lose precision
• Ignoring size and precision needs