Green Go Reads - Serverless DevOps

This inaugural volume of Green Go Reads is about a recent ebook called, “Serverless DevOps: What do we do when the server goes away?” by Tom McLaughlin. A free copy can be downloaded here.

Tom is the founder of ServerlessOps which provides DevOps transformation services with a specialty focus on AWS serverless technology.

Green Go Reads

ALL ABOARD! Operations engineers are moving up the stack! If you haven’t already noticed Serverless, you might miss the train. While Serverless adoption is still growing and the train hasn’t left the station, this ebook prepares the operations folks that change is on the horizon. Change doesn’t have to be grim, but I believe this foreward thinking allows the pragmatic operations people to consider how the Serverless movement will effect their career. Operations engineers that are open-minded to see the future of application development requires them to evolve and adapt to new technology will benefit from reading this ebook. I know I did.

The ebook starts with setting baselines and uses the AWS definition for Serverless which I tend to agree with. Some people think Lambda is required to make an architecure serverless which is simply not true. I tried to point that out in a previous post about making this Serverless blog.

AWS describes Serverless with these four characteristics:

  • No servers to manage or provision
  • Priced by consumption (not capacity)
  • Scales with usage
  • Availability and fault tolerance built in

As an operations engineer looking at these characteristics, you might be concerned because these directly impact your duties. But, who cares? Is there actually operations engineers out there that like the stress of managing or provisioning servers? This is a chance to focus on something new and leave that headache or undifferentiated heavy lifting to cloud providers.

The ebook goes on to explain the evolution of DevOps which brought developers and operations closer to achieve a more efficient process. With Serverless, the responsibility of operations is impacted because the lower levels of the stack are abstracted away by the cloud provider. Probably the most important takeway for me is when the ebook wrote, "There's no avoiding the need for operations engineers to learn to code."

To be honest, I saw this coming after the first time I used the Serverless Framework and Lambda. It was reassuring to read how important it will be for operations to move closer to the codebase aka “up the stack”. And it makes sense, right? Since operations isn’t required to make sure the underlying infrastructure is adequetly performing, the code running the application becomes the critical piece. I like how the ebook points out that you don’t need to be a software engineer to help fix minor bugs.

I started learning Python a while back which has been a simple language for me to quickly dive into code and understand what’s going on. I also have Go and Javascript in the queue once I’m comfortable with Python. Go is another simple yet powerful language built by Google. Go is a compiled language so it’s fast and scalable because it can run simultaneous tasks within an application. Javascript is extremely popular because it’s also fast on the client-side. Eventually, I’d like to get into Vue.js which is a frontend Javascript framework but it’s more important to learn the fundamentals first.

Reading about the roles and responsibilities of operations that change in a Serverless architecture had me excited! I know some people will be worried about job security but this movement just seems like an opportunity to learn something new. The skills as operations engineers are still applicable but the patience to learn something new will be deciding.

Share this post: