Update README.MD
[poolifier.git] / README.MD
1 # Node Thread Pool :arrow_double_up: :on:
2 [![JavaScript Style Guide](https://img.shields.io/badge/code_style-standard-brightgreen.svg)](https://standardjs.com)
3 [![Dependabot](https://badgen.net/dependabot/dependabot/dependabot-core/?icon=dependabot)](https://badgen.net/dependabot/dependabot/dependabot-core/?icon=dependabot)
4 [![Actions Status](https://github.com/pioardi/node-pool/workflows/NodeCI/badge.svg)](https://github.com/pioardi/node-pool/actions)
5
6 <h2>Contents </h2>
7 <h3 align="center">
8 <a href="#installation">Installation</a>
9 <span> · </span>
10 <a href="#usage">Usage</a>
11 <span> · </span>
12 <a href="#api">API</a>
13 <span> · </span>
14 <a href="#cyp">Choose a pool</a>
15 <span> · </span>
16 <a href="#contribute">Contribute</a>
17 <span> · </span>
18 <a href="#nv">Compatibility</a>
19 <span> · </span>
20 <a href="#license">License</a>
21 </h3>
22
23 <h2> Overview </h2>
24 Node pool contains two <a href="https://nodejs.org/api/worker_threads.html#worker_threads_worker_threads">worker-threads </a> pool implementations , you don' t have to deal with worker-threads complexity. <br>
25 The first implementation is a static thread pool , with a defined number of threads that are started at creation time and will be reused.<br>
26 The second implementation is a dynamic thread pool with a number of threads started at creation time ( these threads will be always active and reused) and other threads created when the load will increase ( with an upper limit ), the new created threads will be stopped after a configurable period of inactivity. <br>
27 You have to implement your worker extending the ThreadWorker class<br>
28 <h2 id="installation">Installation</h2>
29
30 ```
31 npm install node-thread-pool --save
32 ```
33 <h2 id="usage">Usage</h2>
34
35 You can implement a worker in a simple way , extending the class ThreadWorker :
36
37 ```js
38 'use strict'
39 const { ThreadWorker } = require('node-pool')
40
41 class MyWorker extends ThreadWorker {
42 constructor () {
43 super((data) => {
44 // this will be executed in the worker thread,
45 // the data will be received by using the execute method
46 return { ok: 1 }
47 }, { maxInactiveTime: 1000 * 60})
48 }
49 }
50 module.exports = new MyWorker()
51 ```
52
53 Instantiate your pool based on your needed :
54
55 ```js
56 'use strict'
57 const { FixedThreadPool, DynamicThreadPool } = require('node-pool')
58
59 // a fixed thread pool
60 const pool = new FixedThreadPool(15,
61 './yourWorker.js')
62
63 // or a dynamic thread pool
64 const pool = new DynamicThreadPool(10, 100,
65 './yourWorker.js')
66 pool.emitter.on('FullPool', () => console.log('Pool is full'))
67
68 // the execute method signature is the same for both implementations,
69 // so you can easy switch from one to another
70 pool.execute({}).then(res => {
71 console.log(res)
72 }).catch ....
73
74 ```
75
76 <strong> See examples folder for more details.</strong>
77
78 <h2 id="nv">Node versions</h2>
79
80 You can use node version 10.x with --experimental-worker flag, or you can use an higher version (i.e 12.x) <br>
81
82 <h2 id="api">API</h2>
83
84 ### `pool = new FixedThreadPool(numThreads, filePath, opts)`
85 `numThreads` (mandatory) Num of threads for this worker pool <br>
86 `filePath` (mandatory) Path to a file with a worker implementation <br>
87 `opts` (optional) An object with these properties :
88 - `errorHandler` - A function that will listen for error event on each worker thread
89 - `onlineHandler` - A function that will listen for online event on each worker thread
90 - `exitHandler` - A function that will listen for exit event on each worker thread
91 - `maxTasks` - This is just to avoid not useful warnings message, is used to set <a href="https://nodejs.org/dist/latest-v12.x/docs/api/events.html#events_emitter_setmaxlisteners_n">maxListeners</a> on event emitters ( workers are event emitters)
92
93 ### `pool = new DynamicThreadPool(min, max, filePath, opts)`
94 `min` (mandatory) Same as FixedThreadPool numThreads , this number of threads will be always active <br>
95 `max` (mandatory) Max number of workers that this pool can contain, the new created threads will die after a threshold ( default is 1 minute , you can override it in your worker implementation). <br>
96 `filePath` (mandatory) Same as FixedThreadPool <br>
97 `opts` (optional) Same as FixedThreadPool <br>
98
99 ### `pool.execute(data)`
100 Execute method is available on both pool implementations ( return type : Promise): <br>
101 `data` (mandatory) An object that you want to pass to your worker implementation <br>
102
103 ### `pool.destroy()`
104 Destroy method is available on both pool implementations.<br>
105 This method will call the terminate method on each worker.
106
107
108 ### `class YourWorker extends ThreadWorker`
109 `fn` (mandatory) The function that you want to execute on the worker thread <br>
110 `opts` (optional) An object with these properties :
111 - `maxInactiveTime` - Max time to wait tasks to work on ( in ms) , after this period the new worker threads will die.
112
113 <h2 id="cyp">Choose your pool</h2>
114 Performance is one of the main target of these thread pool implementation, we want to have a strong focus on this.<br>
115 We already have a bench folder where you can find some comparisons.
116 To choose your pool consider that with a FixedThreadPool or a DynamicThreadPool ( in this case is important the min parameter passed to the constructor) your application memory footprint will increase . <br>
117 Increasing the memory footprint your application will be ready to accept more CPU bound tasks, but during idle time your application will consume more memory. <br>
118 One good choose from my point of view is to profile your application using Fixed/Dynamic thread pool , and to see your application metrics when you increase/decrease the num of threads. <br>
119 For example you could keep the memory footprint low choosing a DynamicThreadPool with 5 threads, and allow to create new threads until 50/100 when requests, this is the advantage to use the DynamicThreadPool. <br>
120 But in general , <strong>always profile your application </strong>
121
122 <h2 id="contribute">Contribute</h2>
123
124 See guidelines [CONTRIBUTING](./.github/CONTRIBUTING.md)
125
126
127 <h2 id="license">License</h2>
128
129 [MIT](./LICENSE)
130