An Introduction to Web Workers

An Introduction to Web Workers

Learn how to use web workers in JavaScript to create parallel programming and perform multiple operations simultaneously rather than interleaving them.

Learn how to use web workers in JavaScript to create parallel programming and perform multiple operations simultaneously rather than interleaving them.

Web workers enable developers to benefit from parallel programming in JavaScript. Parallel programming lets us run different computations at the same time. Let’s see how we benefit from doing tasks in parallel as humans to help us understand the value of web workers.

Let’s imagine that each one of us runs a tech blog. We work by ourselves and are responsible for coding a demo app, writing about the app development process, and creating assets for the post (like diagrams, images, or logos). That’s a heavy task pipeline to handle by ourselves. Coding the app and writing the content are interconnected since the experience of writing the code motivates writing the content. However, creating assets is something that we could delegate to someone else.

Let’s say that we have a group of friends that are talented designers. They agree to create the assets for us. All that they ask is for us to message them a description or sketch of the asset and they will reply back with a cool professional version of it when they are done. Now, we only need to focus on coding, writing, and integrating the assets of the designers once they message us.

This will give us a huge productivity boost. Initially, we were being blocked from coding and writing whenever we had to design assets. The completion of a blog post would have taken much longer if we were to work alone than if we were to delegate one task to a designer friend.

The design team handles the task asynchronously on their own pipeline. The design team acts exactly how a web worker acts in JavaScript applications. JavaScript is a single-threaded language. As such, running expensive logic in the main thread can block it and make our JavaScript application seem slow or unresponsive. Using web workers, we can create a separate thread to run any logic without interrupting the main thread.

[“Web workers in JavaScript allows us to create parallel programming to perform multiple operations simultaneously rather than interleaving them.”> [“Web workers in JavaScript allows us to create parallel programming to perform multiple operations simultaneously rather than interleaving them.”
Let’s explore what we need to know to make use of web workers in JavaScript and what benefits it brings to a web application.
[“Web workers in JavaScript allows us to create parallel programming to perform multiple operations simultaneously rather than interleaving them.”> [“Web workers in JavaScript allows us to create parallel programming to perform multiple operations simultaneously rather than interleaving them.”> [“Web workers in JavaScript allows us to create parallel programming to perform multiple operations simultaneously rather than interleaving them.”> [“Web workers in JavaScript allows us to create parallel programming to perform multiple operations simultaneously rather than interleaving them.”## Why use JavaScript Web Workers?

Let’s expand a bit more on what Jason Miller explained in his tweets.

In JavaScript, we can create parallel programming to perform multiple operations simultaneously using web workers. Web workers let us create background threads that are separate from the main execution thread, where we usually run our user interface logic. The core advantage of this workload separation is that we can run expensive operations within an isolated thread without interrupting or affecting the responsiveness and usability of the main thread. When the background thread completes its task it seamlessly notifies the main thread about the results through an event that is managed through regular JavaScript event handling.

[“Web workers in JavaScript allows us to create parallel programming to perform multiple operations simultaneously rather than interleaving them.”> [“Web workers in JavaScript allows us to create parallel programming to perform multiple operations simultaneously rather than interleaving them.”#### Tweet This
[“Web workers in JavaScript allows us to create parallel programming to perform multiple operations simultaneously rather than interleaving them.”> [“Web workers in JavaScript allows us to create parallel programming to perform multiple operations simultaneously rather than interleaving them.”
Web workers effectively enable a form of multi-threading in JavaScript with some restrictions such as not being able to access the DOM and not having access to the web worker’s parent page (the page that created it). With that in mind, let’s learn next how we can create web workers.

Setting Up a Development Environment

Getting hands-on with web workers will help us understand them better! For the purpose of this blog post, we’ll be running the sample code within a CodeSandbox project. It’s easy to bootstrap and run a vanilla JavaScript project there. Please, follow these steps:

<html>

<head>
    <title>Parcel Sandbox</title>
    <meta charset="UTF-8" />
</head>

<body>
    <div id="app"></div>

    <script src="src/main.js"></script>

</body>

</html>

We’ll soon learn why we are creating these files. CodeSandbox uses ParcelJS to bundle the JavaScript application easily.

[“Web workers in JavaScript allows us to create parallel programming to perform multiple operations simultaneously rather than interleaving them.”## Creating Web Workers

To create a web worker, we use the [Worker()](https://developer.mozilla.org/en-US/docs/Web/API/Worker/Worker "Worker()") constructor from the Web Workers API. The Worker() constructor has the following signature:

Worker(aURL, options);


aURL is a string that represents the URL of the script that we want the worker to execute.

options is an object to customize the Worker instance. The allowed options are type, credentials, and name. We don’t need to configure them for the scope of this post.

In practice, we instantiate a web worker in the main thread. The main thread could be represented by a JavaScript file, for example, main.js, that is the entry point to the application. The web worker thread could be represented by another file, for example, worker.js. main.js then creates a new Worker using the worker.js file. Let’s see this in action.

Let’s open src/main.js in our project and populate it with the following code:

// src/main.js

const worker = new Worker("../src/worker.js");


In the code above, worker becomes a Worker instance that will execute the script on worker.js.

[“Web workers in JavaScript allows us to create parallel programming to perform multiple operations simultaneously rather than interleaving them.”
That’s it for the creation of a web worker! We effectively now have two threads available in our application: main and worker. Next, we’ll learn how to communicate between threads.

Sending Messages To and From a Web Worker

In the introduction, we discussed how the internal collaboration between our Content and Design teams at Auth0 resemble the interaction between threads using web workers in JavaScript. In our case, a Content Engineer represents the main thread and the Designer represents the worker thread. How would the main thread ping the worker thread and vice versa? We do that through the [postMessage()](https://developer.mozilla.org/en-US/docs/Web/API/Worker/postMessage "postMessage()") method and the [onmessage](https://developer.mozilla.org/en-US/docs/Web/API/Worker/onmessage "onmessage") event handler from the Web Workers API.

Let’s use the classic Marco Polo game to see this communication in action. In this game, one player shouts “Marco!” and the other player must reply “Polo!”. Within our context we want to do the following:

  1. main.js and worker.js are on standby listening for any message between each other.
  2. main.js sends a message to worker.js: "Marco!".
  3. worker.js gets the message from main.js and replies: "Polo!".
  4. Step 2 and Step 3 are repeated infinitely.

Step 1: Listen for Messages

The Worker.onmessage event handler let us listen for messages between the threads. The signature of this Worker event handler property is as follows:

myWorker.onmessage = e => {
  // Event handler logic
};


The function assigned to onmessage is called when a message event occurs.

To set this up in main.js, we use the Worker instance we created:

// src/main.js

const worker = new Worker("../src/worker.js");

worker.onmessage = e => {};


To set this up in the web worker thread represented by worker.js, we use the onmessage property directly:

// src/worker.js

onmessage = e => {};


How do we access the message data that is being sent? The message payload can be accessed from the message event’s data property.

Let’s update our code as follows:

// src/main.js

const worker = new Worker("../src/worker.js");

worker.onmessage = e => {
  const message = e.data;
  console.log(`[From Worker]: ${message}`);
};

// src/worker.js

onmessage = e => {
  const message = e.data;
  console.log(`[From Main]: ${message}`);
};


Let’s save our work for each file. On CodeSandbox, we can use CMD + S or CTRL + S to save each file.

We got our threads listening for messages between each other. Next, let’s learn how to send messages.

Step 2: Send a Message from Main Thread to Worker Thread

To send messages, we rely on the Worker.postMessage() method:

worker.postMessage(message);


The postMessage() takes a single parameter representing the data that we want to send. This data may be any value or JavaScript object handled by the structured clone algorithm. As noted by MDN, the structured clone algorithm is an algorithm defined by the HTML5 specification for copying complex JavaScript objects. Why do we need to rely on this algorithm? Data transferred through web workers is passed as a copy, not as a reference.

With an understanding of how postMessage() work, let’s use this method to send a message from the main thread to the worker thread:

// src/main.js

const worker = new Worker("../src/worker.js");

worker.onmessage = e => {
  const message = e.data;
  console.log(`[From Worker]: ${message}`);
};

worker.postMessage("Marco!");


Let’s save our work and open the application preview on its own tab. This can be done by clicking on the Open In New Window button present in the navigation bar of the embedded browser:

In the new preview browser tab, let’s open the browser developer console and refresh the page. We should see the following output:

// [From Main]: Marco!


This output in the console confirms that our web worker is listening and reacting to the message event sent from main.js. Now, we need to reverse the communication. We need to send a message reply from worker.js to main.js.

Step 3: Send a Message from Worker Thread to Main Thread

This will be quick. We need to use the postMessage() method in the onmessage event handler in worker.js:

// src/worker.js

onmessage = e => {
  const message = e.data;
  console.log(`[From Main]: ${message}`);

  postMessage("Polo!");
};


Let’s save our work and refresh the preview browser tab. In the console, we should now see the following output:

// [From Main]: Marco!
// [From Worker]: Polo!


We have achieved bi-directional communication between threads, but the communication is short-lived. Let’s make this multi-threaded Marco Polo game run infinitely.

Step 4: Send Messages Between Main and Worker Infinitely

We are going to keep the communication between threads going endlessly. To better pace the back and forth, we are going to rely on [setTimeout()](https://developer.mozilla.org/en-US/docs/Web/API/WindowOrWorkerGlobalScope/setTimeout "setTimeout()") to delay messaging by 3 seconds.

To start, when main.js gets a message from worker.js, it replies back after 3 seconds:

// src/main.js

const worker = new Worker("../src/worker.js");

worker.onmessage = e => {
  const message = e.data;
  console.log(`[From Worker]: ${message}`);

  const reply = setTimeout(() => worker.postMessage("Marco!"), 3000);
};

worker.postMessage("Marco!");


Next, when worker.js gets a message from main.js it also replies back after 3 seconds:

// src/worker.js

onmessage = e => {
  const message = e.data;
  console.log(`[From Main]: ${message}`);

  const reply = setTimeout(() => postMessage("Polo!"), 3000);
};


The 3 seconds delay creates an eye-friendly pause to be able to see the communication calmly in the developer console. What makes this work infinitely is that every handled message event executes a postMessage() response. Before, the worker.onmessage in main.js did not have a reply within its body.

Let’s save our work and head back to the browser preview. Let’s refresh the page. After a few seconds, we should see the following output:

// [From Main]: Marco!
// [From Worker]: Polo!
// [From Main]: Marco!
// [From Worker]: Polo!


Your browser doesn’t support HTML5 video. Here is a instead.

This will go on forever until we close the browser tab running the preview of our application. But, we could also terminate the web worker manually. Let’s see how we can do that next.

Terminating a Web Worker

We can terminate web workers from the main thread immediately or from the worker thread.

From the main thread, we can terminate a web worker by calling the [terminate()](https://developer.mozilla.org/en-US/docs/Web/API/Worker "terminate()") method of the Web Workers API:

worker.terminate();


After terminate() is issued, the web worker is destroyed immediately without any chance of completing any ongoing or pending operations. The web worker is also given no time to clean up. Thus, terminating a web worker abruptly may lead to memory leaks.

We can also terminate a web worker from the worker thread using its own [close](https://developer.mozilla.org/en-US/docs/Web/API/DedicatedWorkerGlobalScope/close "close") method:

close();


Upon calling close(), any queued tasks present in the event loop are discarded and the web worker scope is closed.

Checking the documentation for close() may be confusing at first because there is a version of the [close()](https://developer.mozilla.org/en-US/docs/Web/API/WorkerGlobalScope "close()") method that has been deprecated. The deprecated version belongs to the WorkerGlobalScope interface. In reality, there are two types of web workers that we can create: dedicated and shared web workers. Each web worker type has its own interface, [DedicatedWorkerGlobalScope](https://developer.mozilla.org/en-US/docs/Web/API/DedicatedWorkerGlobalScope "DedicatedWorkerGlobalScope") and [SharedWorkerGlobalScope](https://developer.mozilla.org/en-US/docs/Web/API/SharedWorkerGlobalScope/close "SharedWorkerGlobalScope") respectively. For the scope of this introduction, we’ve used a dedicated web worker under the hood. The difference between these two types of web workers and how and where to use them will be addressed in a future post along with best practices on terminating workers!

Recap

We’ve learned the basics of how to create a web worker. We learned how to effectively send messages between two threads and how to react to those messages. We briefly touched on the subject of terminating web workers. This last task is to be handled with care and deserves a more detailed explanation. Badly terminated web workers may lead to memory leaks in the application.

[“Web workers in JavaScript allows us to create parallel programming to perform multiple operations simultaneously rather than interleaving them.”
What’s left to learn? A lot! Web workers have been around for a long time and they are great at executing expensive logic. This logic will be much more complex than what we’ve done in this blog post. We’d need to learn topics like handling errors, spawning subworkers, using external libraries, and monitoring web workers using developer tools.

Please let me know in the comments how you liked this introduction to web workers and what else you’d like to learn about this handy technology that lets us perform parallel programming in JavaScript.

Mobile App Development Company India | Ecommerce Web Development Company India

Mobile App Development Company India | Ecommerce Web Development Company India

Best Mobile App Development Company India, WebClues Global is one of the leading web and mobile app development company. Our team offers complete IT solutions including Cross-Platform App Development, CMS & E-Commerce, and UI/UX Design.

We are custom eCommerce Development Company working with all types of industry verticals and providing them end-to-end solutions for their eCommerce store development.

Know more about Top E-Commerce Web Development Company

JavaScript developers should you be using Web Workers?

JavaScript developers should you be using Web Workers?

Do you think JavaScript developers should be making more use of Web Workers to shift execution off of the main thread?

Originally published by David Gilbertson at https://medium.com

So, Web Workers. Those wonderful little critters that allow us to execute JavaScript off the main thread.

Also known as “no, you’re thinking of Service Workers”.

Photo by Caleb Jones on Unsplash

Before I get into the meat of the article, please sit for a lesson in how computers work:

Understood? Good.

For the red/green colourblind, let me explain. While a CPU is doing one thing, it can’t be doing another thing, which means you can’t sort a big array while a user scrolls the screen.

This is bad, if you have a big array and users with fingers.

Enter, Web Workers. These split open the atomic concept of a ‘CPU’ and allow us to think in terms of threads. We can use one thread to handle user-facing work like touch events and rendering the UI, and different threads to carry out all other work.

Check that out, the main thread is green the whole way through, ready to receive and respond to the gentle caress of a user.

You’re excited (I can tell), if we only have UI code on the main thread and all other code can go in a worker, things are going to be amazing (said the way Oprah would say it).

But cool your jets for just a moment, because websites are mostly about the UI — it’s why we have screens. And a lot of a user’s interactions with your site will be tapping on the screen, waiting for a response, reading, tapping, looking, reading, and so on.

So we can’t just say “here’s some JS that takes 20ms to run, chuck it on a thread”, we must think about where that execution time exists in the user’s world of tap, read, look, read, tap…

I like to boil this down to one specific question:

Is the user waiting anyway?

Imagine we have created some sort of git-repository-hosting website that shows all sorts of things about a repository. We have a cool feature called ‘issues’. A user can even click an ‘issues’ tab in our website to see a list of all issues relating to the repository. Groundbreaking!

When our users click this issues tab, the site is going to fetch the issue data, process it in some way — perhaps sort, or format dates, or work out which icon to show — then render the UI.

Inside the user’s computer, that’ll look exactly like this.

Look at that processing stage, locking up the main thread even though it has nothing to do with the UI! That’s terrible, in theory.

But think about what the human is actually doing at this point. They’re waiting for the common trio of network/process/render; just sittin’ around with less to do than the Bolivian Navy.

Because we care about our users, we show a loading indicator to let them know we’ve received their request and are working on it — putting the human in a ‘waiting’ state. Let’s add that to the diagram.

Now that we have a human in the picture, we can mix in a Web Worker and think about the impact it will have on their life:

Hmmm.

First thing to note is that we’re not doing anything in parallel. We need the data from the network before we process it, and we need to process the data before we can render the UI. The elapsed time doesn’t change.

(BTW, the time involved in moving data to a Web Worker and back is negligible: 1ms per 100 KB is a decent rule of thumb.)

So we can move work off the main thread and have a page that is responsive during that time, but to what end? If our user is sitting there looking at a spinner for 600ms, have we enriched their experience by having a responsive screen for the middle third?

No.

I’ve fudged these diagrams a little bit to make them the gorgeous specimens of graphic design that they are, but they’re not really to scale.

When responding to a user request, you’ll find that the network and DOM-manipulating part of any given task take much, much longer than the pure-JS data processing part.

I saw an article recently making the case that updating a Redux store was a good candidate for Web Workers because it’s not UI work (and non-UI work doesn’t belong on the main thread).

Chucking the data processing over to a worker thread sounds sensible, but the idea struck me as a little, umm, academic.

First, let’s split instances of ‘updating a store’ into two categories:

  1. Updating a store in response to a user interaction, then updating the UI in response to the data change
  2. Not that first one

If the first scenario, a user taps a button on the screen — perhaps to change the sort order of a list. The store updates, and this results in a re-rendering of the DOM (since that’s the point of a store).

Let me just delete one thing from the previous diagram:

In my experience, it is rare that the store-updating step goes beyond a few dozen milliseconds, and is generally followed by ten times that in DOM updating, layout, and paint. If I’ve got a site that’s taking longer than this, I’d be asking questions about why I have so much data in the browser and so much DOM, rather than on which thread I should do my processing.

So the question we’re faced with is the same one from above: the user tapped something on the screen, we’re going to work on that request for hopefully less than a second, why would we want to make the screen responsive during that time?

OK what about the second scenario, where a store update isn’t in response to a user interaction? Performing an auto-save, for example — there’s nothing more annoying than an app becoming unresponsive doing something you didn’t ask it to do.

Actually there’s heaps of things more annoying than that. Teens, for example.

Anyhoo, if you’re doing an auto-save and taking 100ms to process data client-side before sending it off to a server, then you should absolutely use a Web Worker.

In fact, any ‘background’ task that the user hasn’t asked for, or isn’t waiting for, is a good candidate for moving to a Web Worker.

The matter of value

Complexity is expensive, and implementing Web Workers ain’t cheap.

If you’re using a bundler — and you are — you’ll have a lot of reading to do, and probably npm packages to install. If you’ve got a create-react-app app, prepare to eject (and put aside two days twice a year to update 30 different packages when the next version of Babel/Redux/React/ESLint comes out).

Also, if you want to share anything fancier than plain data between a worker and the main thread you’ve got some more reading to do (comlink is your friend).

What I’m getting at is this: if the benefit is real, but minimal, then you’ve gotta ask if there’s something else you could spend a day or two on with a greater benefit to your users.

This thinking is true of everything, of course, but I’ve found that Web Workers have a particularly poor benefit-to-effort ratio.

Hey David, why you hate Web Workers so bad?

Good question.

This is a doweling jig:

I own a doweling jig. I love my doweling jig. If I need to drill a hole into the end of a piece of wood and ensure that it’s perfectly perpendicular to the surface, I use my doweling jig.

But I don’t use it to eat breakfast. For that I use a spoon.

Four years ago I was working on some fancy animations. They looked slick on a fast device, but janky on a slow one. So I wrote fireball-js, which executes a rudimentary performance benchmark on the user’s device and returns a score, allowing me to run my animations only on devices that would render them smoothly.

Where’s the best spot to run some CPU intensive code that the user didn’t request? On a different thread, of course. A Web Worker was the correct tool for the job.

Fast forward to 2019 and you’ll find me writing a routing algorithm for a mapping application. This requires parsing a big fat GeoJSON map into a collection of nodes and edges, to be used when a user asks for directions. The processing isn’t in response to a user request and the user isn’t waiting on it. And so, a Web Worker is the correct tool for the job.

It was only when doing this that it dawned on me: in the intervening quartet of years, I have seen exactly zero other instances where Web Workers would have improved the user experience.

Contrast this with a recent resurgence in Web Worker wonderment, and combine that contrast with the fact that I couldn’t think of anything else to write about, then concatenate that combined contrast with my contrarian character and you’ve got yourself a blog post telling you that maybe Web Workers are a teeny-tiny bit overhyped.

Thanks for reading

If you liked this post, share it with all of your programming buddies!

Follow us on Facebook | Twitter

Further reading

An Introduction to Web Workers

JavaScript Web Workers: A Beginner’s Guide

Using Web Workers to Real-time Processing

How to use Web Workers in Angular app

Using Web Workers with Angular CLI


Hire PHP Developer and Web Developer for your Online Business

Hire PHP Developer and Web Developer for your Online Business

PHP is widely used open-source scripting language it helps in making dynamically easy your websites and web application. Mobiweb Technology is your best technical partner and offering you solution for any kind of website and application...

PHP is widely used open-source scripting language it helps in making dynamically easy your websites and web application. Mobiweb Technology is your best technical partner and offering you solution for any kind of website and application development. To hire PHP developer and web developer at affordable prices contact Mobiweb Technology via [email protected]