How Big Is WebAssembly?
When WebAssembly debuted circa 2015 as a way to streamline the creation and execution of applications within browsers, it was pounced on by developers. Since it converts programming languages such as C, C++, JavaScript, or Rust to a binary-code format that fits the Web, its potential for use in browser-based apps seemed ideal. Indeed, WebAssembly (nicknamed Wasm by proponents) has made it into a range of commercial applications, including ones from Cloudflare, Adobe, Fastly, Shopify, Capital One, VMware, and Sony, to name just a few.
Now, Wasm is at a turning point, heading into a bright future as a tool for a number of uses, including functions in serverless computing, security, Internet of Things (IoT), and edge computing environments. Wasm’s sleek runtime profile looks to make it a strategic element in the toolbox of cloud developers.
Let’s take a closer look at how that’s unfolding.
Beyond the Browser
While the basic specs of Wasm were initially shaped by the World Wide Web Consortium (W3C), Mozilla in 2019 started a project to extend Wasm from browsers to server operating systems. Dubbed the WebAssembly System Interface (WASI), the project was taken up by a group called the Bytecode Alliance. Members include representatives of Amazon, Arm, Cisco, Docker, Fastly, GitHub, Intel, Microsoft, Mozilla, and VMware, among others. Companies dedicated to commercial Wasm platforms, including startups Cosmonic and Fermyon, also are members.
In 2020, Wasm technology was incorporated into another project called WasmCloud undertaken by the Cloud Native Computing Foundation (CNCF). The purpose, according to the group’s website, is to “Write your components once and run them anywhere at any scale [emphasis in original]. Deploy to the cloud, the edge, browsers, IoT, or anywhere in between.” Key to this project is using Wasm as a replacement for as well as an accompaniment to Kubernetes and virtual machines.
The Wasm news doesn’t stop there. Circa 2021 a new effort began to “modularize” Wasm. The basic goal is to create a WebAssembly Component Model, comprising modules that support different languages and functions, which can be fastened together in “Lego”-like fashion, building on Wasm’s compiler-like capabilities.
A key point here is that Wasm modules can be used to create applications capable of running in different environments – in browsers, in clouds, in IoT devices, etc. – without those apps having to be rewritten from the ground up for each location. Today, Wasm work is focused heavily on the Wasm Component Model and its application to today’s evolving distributed, hybrid environments.
Beyond the Tech Talk
Any talk of Wasm quickly leads into the technological weeds. But what is the high-level view? The smart money seems to think the WebAssembly Component Model could lead the way to more efficient serverless computing, as well as to faster and more secure edge computing and IoT apps.
“Wasm offers a lightweight runtime for serverless,” said Matt Butcher, CEO and co-founder of Fermyon, which creates a platform-as-a-service based on WebAssembly. There’s also a security component to Wasm’s Component Model, he maintains. That’s because Wasm runs code in a sandbox format, meaning that key functions of a program are sealed off from other areas in a host environment that could allow malefactors to retrieve information.
Downside and Upsides
Security, debugging, IoT, serverless, edge computing – all seem to be set to benefit from Wasm. So what’s the holdup?
"I'm a firm believer in the technology overall, but progress has been slower than I'd hoped," said Colin Murphy, senior software engineer, CC Web at Adobe. "We are in many ways creating a new operating system, so there is just a lot of work that needs to be done. There are a lot of things Wasm can do right now very well, but it will be a very long time before a significant percentage of existing real life applications can be compiled to Wasm. There are however many opportunities that people can take advantage of right now and the community's momentum is increasing."
Fermyon’s CEO Matt Butcher thinks Wasm is part of the tool chain for a variety of compute platforms and that it has a bright future. His company’s progress seems to highlight that view. Since 2021, Fermyon has garnered $26 million in seed and Series A funding from Amplify Partners and Insight Partners, along with a slew of independent investors that include Tyler McMullen (the CTO of Fastly), Grant Miller and Marc Campbell (CEO and CTO, respectively, at Replicated), Armon Dadgar (co-founder and CTO of HashiCorp), Daniel Lopez Ridruejo (co-founder and former CEO of Bitnami now part of VMware), and Lachlan Evenson (principal program manager on the open source team at Microsoft Azure). The firm has about 30 employees and has released products called Fermyon Spin and Fermyon Cloud that enable developers to create WebAssembly applications today.
In a press release announcing the Series A, Butcher stated:
“WebAssembly is a ‘build once, run anywhere’ technology designed to enable apps to run inside the browser that is now being applied to solving cloud problems where speed and security are paramount. It has quickly become a foundation for the next wave of cloud computing alongside the container and virtual machine technologies already in use today. “
While it may take time to fully mature, it looks as though WebAssembly is on the rise among developers looking to simplify the creation and maintenance of cloud applications. What started in the browser is spreading steadily outward.
