Benchmark: Array.find vs. Map.get for 1000 - 1e6

Array.find vs. Map.get for 1000 - 1e6 (version: 1)

Test the performance of Array.find vs. Map.get, with various array sizes, and with/without conversion from array to map

Comparing performance of: `Array.find, 1000 elements vs Map.get, 1000 elements vs Array.find, 1 000 000 elements vs Map.get, 1 000 000 elements vs Array to Map + Map.get, 1 lookup vs Array to Map + Map.get, 10 lookups vs Array to Map + Map.get, 1000 lookups vs Array.find, 10 lookups vs Array.find, 1000 lookups`

Created: one year ago by: Guest

Go to the latest result

Script Preparation code:

function getRandomElement(id) {
  return {
    id,
    a: Math.random(),
    b: Math.random(),
    c: Math.random(),
  }
}

function getArray(length) {
  const result = [];
  for (let i = 0; i < length; i++) {
    result.push(getRandomElement(i))
  }
  
  return result;
}

function arrayToMap(array) {
  return new Map(array.map(el => [el.id, el]));
}

function getRandomInt(max) {
  return Math.floor(Math.random() * max);
}

array_small = getArray(1000);
array_large = getArray(1000000);

map_small = arrayToMap(array_small);
map_large = arrayToMap(array_large);

Tests:

Array.find, 1000 elements
const target = getRandomInt(99); array_small.find(el => el.id === target);
Map.get, 1000 elements
const target = getRandomInt(99); map_small.get(target);
Array.find, 1 000 000 elements
const target = getRandomInt(999999); array_large.find(el => el.id === target);
Map.get, 1 000 000 elements
const target = getRandomInt(999999); map_large.get(target)
Array to Map + Map.get, 1 lookup
const map = arrayToMap(array_small); const target = getRandomInt(9999); map.get(target);
Array to Map + Map.get, 10 lookups
const map = arrayToMap(array_small); for (let i = 0; i < 10; i++) { const target = getRandomInt(9999); map.get(target); }
Array to Map + Map.get, 1000 lookups
const map = arrayToMap(array_small); for (let i = 0; i < 1000; i++) { const target = getRandomInt(9999); map.get(target); }
Array.find, 10 lookups
for (let i = 0; i < 10; i++) { const target = getRandomInt(9999); array_small.find(el => el.id === target) }
Array.find, 1000 lookups
for (let i = 0; i < 1000; i++) { const target = getRandomInt(9999); array_small.find(el => el.id === target) }

Rendered benchmark preparation results:

Suite status: <idle, ready to run>

Previous results

Test case name	Result
Array.find, 1000 elements
Map.get, 1000 elements
Array.find, 1 000 000 elements
Map.get, 1 000 000 elements
Array to Map + Map.get, 1 lookup
Array to Map + Map.get, 10 lookups
Array to Map + Map.get, 1000 lookups
Array.find, 10 lookups
Array.find, 1000 lookups

Fastest: N/A

Slowest: N/A

Latest run results:

Run details: (Test run date: one year ago)

User agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/132.0.0.0 Safari/537.36

Browser/OS: Chrome 132 on Mac OS X 10.15.7

View result in a separate tab

Map.get, 1000 elements

115.8M/s

Map.get, 1 000 000 elements

111.7M/s

Array.find, 1000 elements

17.0M/s

Array.find, 10 lookups

122K/s

Array to Map + Map.get, 10 lookups

45K/s

Array to Map + Map.get, 1 lookup

41K/s

Array to Map + Map.get, 1000 lookups

38K/s

Array.find, 1000 lookups

1K/s

Array.find, 1 000 000 elements

429/s

View exact numbers

Test name	Executions per second
✓ Map.get, 1000 elements	115,812,648 Ops/sec
Map.get, 1 000 000 elements	111,740,784 Ops/sec
Array.find, 1000 elements	17,040,412 Ops/sec
Array.find, 10 lookups	121,601 Ops/sec
Array to Map + Map.get, 10 lookups	44,901 Ops/sec
Array to Map + Map.get, 1 lookup	40,776 Ops/sec
Array to Map + Map.get, 1000 lookups	38,473 Ops/sec
Array.find, 1000 lookups	1,201 Ops/sec
Array.find, 1 000 000 elements	429 Ops/sec

Autogenerated LLM Summary (model gpt-4o-mini, generated one year ago):

LLMs can make mistakes. Check important info.

The benchmark compares two different approaches for retrieving elements from collections in JavaScript: using `Array.find` on an array and using `Map.get` on a Map. Both methods are evaluated under varying conditions, specifically different input sizes (1000 elements and 1,000,000 elements) and the additional overhead of converting an array to a map.

### Key Comparisons and Test Cases

1. **Array.find vs. Map.get**: 
   - **Array.find**: A method that searches through an array for the first element that matches a condition, specified by a callback function. It has a time complexity of O(n), as it potentially checks each element.
   - **Map.get**: A method that retrieves a value from a Map collection based on a specified key. This method provides average time complexity of O(1) due to the underlying hash table structure.

2. **Test Cases**:
   - **Test Case 1**: `Array.find` with 1000 elements.
   - **Test Case 2**: `Map.get` with 1000 elements.
   - **Test Case 3**: `Array.find` with 1,000,000 elements.
   - **Test Case 4**: `Map.get` with 1,000,000 elements.
   - **Test Case 5**: Conversion of small array to Map to perform a single lookup with `Map.get`.
   - **Test Cases 6, 7**: Multiple lookups using the converted Map.
   - **Test Case 8**: Multiple lookups using `Array.find` on the original array.

### Pros and Cons

- **Array.find**:
  - **Pros**:
    - Easy to use and understand.
    - Works with JavaScript’s existing array structure without any additional conversions.
  - **Cons**:
    - Performance degrades with large datasets, especially as the time complexity is linear. This means that for larger arrays, it could become significantly slower.
  
- **Map.get**:
  - **Pros**:
    - Fast access time (constant time complexity) for lookups, making it highly efficient, especially for larger datasets.
    - More robust when frequent lookups are required, thanks to its efficient structure.
  - **Cons**:
    - Requires an upfront conversion of data from an array to a Map, making it potentially less straightforward in situations where only a few lookups are needed.

### Other Considerations

- **Map Conversion Overhead**: Converting an array to a Map introduces a one-time performance cost, which can be justified if multiple lookups follow. If an array is only searched once or twice, it may not be worth the conversion overhead.
  
- **Data Structure Usage**: Choosing between standard arrays and Maps can depend on the requirements of the application. For applications where you have a large set of unique identifiers and need frequent lookups, Maps are generally preferable.

### Alternative Approaches

- **Using Objects**: In some cases, JavaScript Objects could be used as hash tables, where keys are unique IDs, allowing for similar quick lookups. However, this has its limitations regarding prototype properties and non-string keys.
  
- **Data Libraries**: Libraries like Lodash provide utility functions for collections and could offer optimized approaches to searching through datasets. They can often provide built-in methods for searching that can simplify code while potentially improving performance.

- **Set for Uniqueness**: If uniqueness is essential, employing a `Set` might also be considered, but it would not directly assist with key-value pairs like `Map`.

This benchmark illustrates the critical trade-offs in performance between using structured data (Map) versus unstructured (Array) and highlights the importance of choosing the right data structures for the problem at hand.