Serverless

Serverless architectures have very much changed how modern applications are designed and built. They are the ultimate manifestation of the “use only what we need” mindset—we only pay for exactly the work that needs to get done, as opposed to having servers waiting around in case we have work to be done. Many organizations have reduced their costs by up to 90% through rearchitecting to a more cloud-native serverless model.

Of course, serverless architectures are not for every use case. Yes, there are still servers behind those serverless applications, but the great part is that we only need to concern ourselves with the ultimate task rather than all the supporting infrastructure. Generally, serverless options excel in situations where our workload can be broken down into small, discrete steps. That said, a decent portion of workloads can be broken down into smaller steps, which is why serverless architectures are so versatile.

Serverless services can take many shapes, but we will discuss a few common AWS services here. Indeed, serverless architectures are a whole course unto themselves, so we will not go too deep—but we should still be familiar with our options.

AWS Lambda

AWS Lambda is probably the most recognized serverless option on the AWS platform. The sole purpose of Lambda is to run code. All the infrastructure, backups, scaling, redundancy, and logging are handled automatically behind the scenes.

Natively, Lambda supports Node.js, Python, Java, .NET, Go, and Ruby, and if your preferred language is not included, you can ...