What is serverless computing?

To understand what serverless computing is all about and how it came to be, we first need to travel back in time to the era of mainframes and traditional data centers! Sounds like a long-lost time, right? Don't worry, we are not going that far back. This is probably in the time when most IT organizations had massive in-house data centers for hosting almost all enterprise applications and services. First, these applications were hosted directly on physical servers, and then eventually migrated onto the virtualized environments that provided a better utilization of resources as well as helped to slash down the overall costs and time for deployments from months to days. With the advent of the virtualization era, we also started to develop and use more convenient deployment tools that helped to deploy our applications with more ease, but it still meant managing the application's underlying operating system, software platform, and so on:

With virtualization clearly not having all the answers, we started looking for a much simpler application deployment model and, in return, found Containers. Unlike their earlier counterparts, Virtual Machines, Containers don't require a lot of resources or overhead to run. They are far easier and quicker to deploy, hence, reduce the overall application deployment time from days to minutes! You could now easily roll out a new patch for your application, scale your application dynamically based on incoming requests, and even orchestrate various other functions using a vast variety of container management products. However, the question of managing the Containers still remains, and trust me, managing a fleet of thousands of Containers and their underlying physical servers is no easy task. A better, more efficient, deployment model was needed; something that provided us with the agility and flexibility of containers, but without all the hassle and trouble of managing them. Enter serverless computing!

Serverless computing is all about running your application code on small ingots of some CPU and memory without having to worry about the OS type, the software platform, or any of the underlying hardware either. Just take your code and run it! Yes! It's that simple! Serverless computing today is offered by most public cloud providers, such as Amazon Web Services, Google Cloud Platform, Microsoft Azure, and even by IBM as a managed service. This essentially means that all you need to do is write your code or functions that perform a very specific task, select the quantity of resources (in this case RAM) required to run your code and submit it to the serverless cloud computing service to execute on. The service makes sure that your code gets the required amount of memory and CPU cycles it needs to execute. Hence, the collective term Function as a Service (FaaS).