Benchmark: Push vs Spread vs Double loop Ultimate

Script Preparation code:

  var iterations = 50000;

  var a = [];

  for (let i = 0; i < iterations; i++) {
    a.push(Math.round(Math.random() * iterations));
  }

Tests:

Reduce and push
const result = a.reduce((acc, item, index) => { if(item % 2 === 0) { acc.push({ [index]: item }); } return acc}, []); console.log("Reduce and push", result.length);
Reduce and spread
const result = a.reduce((acc, item, index) => (item % 2 === 0 ? [...acc, { [index]: item }] : acc), []); console.log("Reduce and spread", result.length);
Filter and map
const result = a.filter(item => item % 2 === 0).map((oddItem, index) => ({ [index]: oddItem })); console.log("Filter and map", result.length);

Rendered benchmark preparation results:

Suite status: <idle, ready to run>

Previous results

Test case name	Result
Reduce and push
Reduce and spread
Filter and map

Fastest: N/A

Slowest: N/A

Autogenerated LLM Summary (model llama3.2:3b, generated one month ago):

LLMs can make mistakes. Check important info.

**What is being tested?**

The provided JSON represents a JavaScript microbenchmark, which measures the performance of three different approaches for filtering and processing an array:

1. `Reduce and push`
2. `Reduce and spread`
3. `Filter and map`

Each test case uses a similar preparation code to create an array `a` with 50,000 random integers, and then applies one of the three approaches to process the array.

**Options being compared**

The options being compared are:

1. **Reduce and push**: Uses the `reduce()` method with a callback function that pushes elements to an accumulator array (`acc`) if the current element is even.
2. **Reduce and spread**: Uses the `reduce()` method with a callback function that spreads elements from an accumulator array (`acc`) if the current element is odd, or returns the accumulator array unchanged if it's even.
3. **Filter and map**: Uses the `filter()` method to filter out odd numbers, followed by the `map()` method to create a new array with index-value pairs.

**Pros and cons of each approach**

1. **Reduce and push**:
	* Pros: Can be more efficient if the accumulator array is large, as it avoids creating multiple arrays.
	* Cons: May lead to memory issues if the accumulator array grows too large, and can be slower for very large inputs due to the overhead of pushing elements.
2. **Reduce and spread**:
	* Pros: Can be faster for larger inputs, as it avoids creating a new accumulator array and uses the existing one instead.
	* Cons: May lead to performance issues if the accumulator array is already large or if there are many iterations, due to the overhead of spreading elements.
3. **Filter and map**:
	* Pros: Can be more efficient for small inputs, as it avoids the overhead of reducing and pushing/spreading elements.
	* Cons: May lead to slower performance for larger inputs, due to the overhead of filtering and mapping elements.

**Libraries used**

None of the provided code snippets use any external libraries or frameworks. The `reduce()` method is a built-in JavaScript method that creates an accumulator array by iteratively applying a callback function to each element in the original array.

**Special JS features/syntax**

The code snippet uses the following special features:

* **Template literals**: Used for creating multi-line strings (`" \r\n\r\n  const result = ..."`).
* **Arrow functions**: Used as callback functions for `reduce()` and `map()` methods (e.g., `(item, index) => {...}`).

**Alternatives**

Other approaches to process the array could include:

1. Using a loop with manual indexing (e.g., `for (let i = 0; i < iterations; i++) { if (a[i] % 2 === 0) { result.push(a[i]); } }`).
2. Using the `forEach()` method instead of `reduce()`.
3. Using a library like Lodash or Ramda, which provides additional functionality for working with arrays.

Keep in mind that these alternatives may not provide significant performance improvements over the current approaches and may have additional overhead due to the use of libraries or manual indexing.

Latest run results:

Run details: (Test run date: 2 years ago)

User agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/103.0.0.0 Safari/537.36

Browser/OS: Chrome 103 on Windows

View result in a separate tab

Test name	Executions per second
Reduce and push	124.6 Ops/sec
Reduce and spread	2.2 Ops/sec
Filter and map	132.6 Ops/sec

Related benchmarks:

Push vs Spread vs Double loop Ultimate (version: 0)

lets see

Comparing performance of: Reduce and push vs Reduce and spread vs Filter and map

Created: 2 years ago by: Guest

Jump to the latest result

Reduce and push

Reduce and spread

Filter and map

Suite status: <idle, ready to run>

Experimental features:

Fastest: N/A

Slowest: N/A

Autogenerated LLM Summary (model llama3.2:3b, generated one month ago):