What is Serverless Computing?

What if government developers could spend all their time building and deploying great apps, and none of their time managing servers? This may sound too good to be true, especially for agencies that are tasked with managing IT infrastructure on premise.

Rather than focusing solely on their mission — whether that’s public safety or administering benefits — those agencies also have to worry about ongoing maintenance of servers and if those IT resources will be available when needed.

The more time agencies spend provisioning and managing servers, the less time they have for mission-focused activities. To address this issue, agencies began investing in cloud computing – specifically Infrastructure-as-a-Service.

While initial Infrastructure-as-a-Service capabilities provided significant benefits, such as cost savings and improved security, there are still untapped advantages of cloud that agencies have not fully realized. One of those advantages gaining momentum is serverless computing.

Serverless computing is the natural evolution of cloud computing.

At its core, serverless is the abstraction of servers, infrastructure, and operating systems. It’s where cloud providers make server management and capacity planning invisible to developers. This means government employees can focus on innovating and delivering apps faster instead of worrying about infrastructure.

Before we dive into all that serverless entails, it helps to understand how apps have historically been built and the evolution to current compute options — including serverless.

The first phase is on-premise servers. This entails agencies owning servers — and other hardware — installing their own software, and hosting data centers. Maintenance of the technology is complex, resource intensive, and costly. And it’s all on the agency to execute.

Agencies have to consider myriad issues, including the size and number of servers, what operating systems to use, if their servers have the latest security patches, physical security of the facility where the servers are housed, and more.

Infrastructure-as-a-Service is the next phase after on-premise. And it marks the first move to cloud. Think of it this way: With Infrastructure-as-a-Service, you’re essentially renting virtual machines — or servers — from a cloud provider. This means vendors are tasked with maintaining the infrastructure and hardware, as well as responding to any failures.

While Infrastructure-as-a-Service removes the burden of hardware management for agencies, there are still many things to worry about – like deciding how many servers to rent, when to patch them, and which operating system to use.

Next up is Platform-as-a-Service. In this phase, agencies don’t have to worry about patching or monitoring servers. But they still have to choose their operating system and the size of their servers. They also have to pay for server capacity even if it’s not fully utilized.

The last phase in this journey is serverless computing. Serverless provides all the benefits of Platform-as-a-Service, but it takes it one step further.

Agencies no longer have to worry about the server itself. They only pay for the computing power they actually use, and they’re able focus on deploying code rather than infrastructure management. There’s no provisioning of servers, no analyzing and no scaling. They also don’t have to forecast infrastructure demands or pay for server capacity that isn’t used.

In our recent 10-minute course, How Serverless Computing Can Simplify Cloud, we dive deeper into the benefits of serverless computing as well as how to get started. Take the free online course here.