It took a while for enterprises to embrace and move full-throttle to cloud computing, but they have with a vengeance—and now serverless computing may just be the Next Big Thing. So what exactly is it?
For developers, serverless computing is development nirvana. It means they don’t have to worry about infrastructure when deploying code; all computing resources—such as managing, provisioning, and maintaining servers—are dynamically provisioned by the cloud provider. Pricing is on a pay-as-you-go model and based on the resources an application actually consumes rather than a set amount of capacity. This reduces “sunk cost–rather than being based on how long an instance is running, regardless of whether it’s doing any meaningful work at that point in time,” explains Owen Rogers, research director of digital at 451 Research, in the 2017 report Economics of Serverless Cloud Computing.
“For operations, serverless saves the hassle of managing the infrastructure, allowing developers to concentrate their skills elsewhere,” Rogers adds.
Serverless computing is also sometimes referred to as function as a service (FaaS). Its pricing model and the fact that management is performed at the level of individual functions rather than virtual machines (VMs) or containers are the two key characteristics that distinguish it from other cloud variants and significantly drive its economic benefits, according to Rogers.
Serverless has become more popular due in part to AWS introducing its own offering, AWS Lambda and Microsoft rolling out its Functions architecture for Azure, which supports several programming languages. Free, serverless tiers are also offered by Google (Cloud Functions), IBM (OpenWhisk) and Microsoft in their public cloud environments, giving enterprises “powerful capability at no charge. This freemium model will stimulate serverless experimentation by developers and operations alike, helping them gain skills and ultimately fueling the growth of serverless services,’’ Rogers notes.
There are many different use cases for serverless computing, including IoT, web-based apps, mobile apps, and chatbots.
Yet, the concept is not sitting well with DevOps teams because serverless computing takes a bite out of what they do and there is “a massive loss of perceived control,” observes Nick Rockwell, CTO of The New York Times, in a blog advocating for the approach.
Serverless architectures “are a huge innovation and will take over and everyone is just scared and why is everyone messing around with containers, it’s stupid,” Rockwell writes.
Instead of putting so much effort into containerization, where applications are broken into smaller, standardized, and reusable pieces, it makes more sense to use serverless for building software, he maintains.
Internal teams do lose some control when serverless computing is employed because of vendor dependencies. As the next generation of tools, serverless computing will enable enterprises to do more with less staff.
But there’s no denying the productivity benefits. “When we get comfortable building mainstream web apps on serverless platforms, the impact to productivity will be really, really big—bigger pound-for-pound than original-flavor cloud,’’ Rockwell says, adding that his goal is for all apps to become serverless by 2019.
Other companies eyeing serverless computing include Netflix, which is planning to use Lambda to build a rule-based self-managing infrastructure and replace inefficient processes to reduce the number of errors and save developers time.
451 Research’s Voice of the Enterprise (VotE): Cloud Transformation, Workloads and Key Projects 2016 survey found that 37% of IT decision-makers were using serverless technology to some degree: 14% were using serverless in production, with 11% currently testing the technology in pilots or development and the remaining 12% in initial discovery phases.
Instead of replacing today’s delivery models, industry experts say you can expect to see serverless computing complement them for both web-scale and enterprise deployments, and continue growing in adoption over the next few years.
Esther Shein is a longtime freelance writer and editor specializing in technology and business. Previously, she was the editor-in-chief of the online technology magazine Datamation, a managing editor at BYTE and a senior writer at eWeek (formerly PC Week).