Benchmark: UUID Test 3 - MeasureThat.net

Script Preparation code:

function oldmeth() {
  function chr4 () {
    return Math.random().toString(16).slice(-4);
  }
  return `${chr4() + chr4()}-${chr4()}-${chr4()}-${chr4()}-${chr4()}${chr4()}${chr4()}`;
}

function oldmethcrypto() {
  function getRandom () {
    return crypto.getRandomValues(new Uint32Array(4)).reduce((prev, current) => `${prev.toString(16)}${current.toString(16)}`);
  }
  
  function update4(uuid) {
    const fourIndx = 12;
    
    return `${uuid.slice(0,12)}4${uuid.slice(13)}`;
  }
  function getEight(c) {
    return(c ^ crypto.getRandomValues(new Uint8Array(1))[0] & 15 >> c / 4).toString(16);
  }
  function update8(uuid) {
    const eightIndx = 16;
    return `${uuid.slice(0,16)}${getEight(8)}${uuid.slice(17)}`;
  }
  
  function addDashes(uuid) {
    // xxxxxxxx-xxxx-4xxx-yxxx-xxxxxxxxxxxx
    return `${uuid.slice(0,8)}-${uuid.slice(8,12)}-${uuid.slice(12,16)}-${uuid.slice(16,20)}-${uuid.slice(20)}`;
  }
  
  let uuid = getRandom();
  
  uuid = update4(uuid);
  
  uuid = update8(uuid);
  
  uuid = addDashes(uuid);
  
  return uuid;
}


function oldmethcrypto2() {
  function getRandom () {
    return crypto.getRandomValues(new Uint32Array(4)).reduce((prev, current) => `${prev.toString(16)}${current.toString(16)}`);
  }

  function getEight(c) {
    return(c ^ crypto.getRandomValues(new Uint8Array(1))[0] & 15 >> c / 4).toString(16);
  }

  let uuid = getRandom();
  return `${uuid.slice(0,8)}-${uuid.slice(8,12)}-4${uuid.slice(13,16)}-${getEight(8)}${uuid.slice(17,20)}-${uuid.slice(20)}`;
}

function newmethrand1() {
  function getRandom (num) {
    let output;
    for(let i = 0; i < Math.ceil(num/8); i++) {
      output = `${output}${Math.random().toString(16).slice(-8)}`;
    }
  
  	return output.slice(num * -1);
  }

  function getEight(c) {
    return(c ^ Math.random().toString(16).slice(-1) & 15 >> c / 4).toString(16);
  }

  let uuid = getRandom(32);
  return `${uuid.slice(0,8)}-${uuid.slice(8,12)}-4${uuid.slice(13,16)}-${getEight(8)}${uuid.slice(17,20)}-${uuid.slice(20)}`;
}

function newmeth() {
  return ([1e7] + -1e3 + -4e3 + -8e3 + -1e11).replace(/[018]/g, c =>
     (c ^ crypto.getRandomValues(new Uint8Array(1))[0] & 15 >> c / 4).toString(16),
  );                                      
}

Tests:

old old
oldmeth();
new
newmeth();
old crypt
oldmethcrypto();
old crypt 2
oldmethcrypto2();
old meth2 with random
newmethrand1();

Rendered benchmark preparation results:

Suite status: <idle, ready to run>

Previous results

Test case name	Result
old old
new
old crypt
old crypt 2
old meth2 with random

Fastest: N/A

Slowest: N/A

Latest run results:

Run details: (Test run date: 6 years ago)

User agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:61.0) Gecko/20100101 Firefox/61.0

Browser/OS: Firefox 61 on Windows

View result in a separate tab

Test name	Executions per second
old old	106844.4 Ops/sec
new	15775.3 Ops/sec
old crypt	111633.8 Ops/sec
old crypt 2	145212.6 Ops/sec
old meth2 with random	113476.1 Ops/sec

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

LLMs can make mistakes. Check important info.

Measuring the performance of JavaScript code is crucial for optimizing and comparing different implementations.

**Benchmark Definition**

The provided JSON represents a benchmark definition, which outlines the characteristics of the test case:

*   Name: "UUID Test 3"
*   Description: "UUID speed validation tests"
*   Script Preparation Code: This section contains four JavaScript functions that generate a UUID (Universally Unique Identifier) in different ways.
    *   `oldmeth`: This function uses the hexadecimal representation of random numbers to create a UUID.
    *   `oldmethcrypto`: This function uses the crypto module to generate a UUID. It first generates four random 32-bit integers, concatenates them, and then updates and adds dashes to create the final UUID.
    *   `oldmethcrypto2`: Similar to `oldmethcrypto`, but with some differences in the update and add-dashes logic.
    *   `newmethrand1` and `newmeth`: These functions use a custom implementation of UUID generation using random numbers.

**Individual Test Cases**

The individual test cases are defined as follows:

*   `old old`: This test case uses the `oldmeth` function to generate a UUID.
*   `new`: This test case uses the `newmeth` function to generate a UUID.
*   `old crypt` and `old crypt 2`: These test cases use the `oldmethcrypto` and `oldmethcrypto2` functions, respectively, to generate UUIDs.
*   `old meth2 with random`: This test case uses the `newmethrand1` function to generate a UUID.

**Latest Benchmark Result**

The latest benchmark result shows the performance of each test case on Firefox 61 running on Windows:

| Test Name | Executions Per Second |
| --- | --- |
| old crypt 2 | 113476.0625 |
| old meth2 with random | 106844.390625 |
| old crypt | 111633.7578125 |
| old old | 15775.28125 |
| new | 145212.59375 |

**Analysis and Observations**

Based on the benchmark results, we can observe:

*   The `oldmeth` function is significantly slower than the other implementations.
*   The `newmeth` function performs better than the `oldmethcrypto2` function.
*   The `old meth2 with random` function has a performance similar to the `newmeth` function.

These observations suggest that the custom implementation of UUID generation in `newmethrand1` and `newmeth` is more efficient than the crypto-based implementations.

Related benchmarks:

UUID Test 3 (version: 2)

UUID speed validation tests

Comparing performance of: old old vs new vs old crypt vs old crypt 2 vs old meth2 with random

Created: 6 years ago by: Registered User

Jump to the latest result

old old

new

old crypt

old crypt 2

old meth2 with random

Suite status: <idle, ready to run>

Experimental features:

Fastest: N/A

Slowest: N/A

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