Off-by-none: Issue #7

October 16, 2018

Livin’ in the (serverless) future 🤖

Welcome to Issue #7 of Off-by-none. I’m excited to have you here! 🙌 Let’s talk about serverless.

Last week we talked about the future of serverless now that institutional investors are starting to get behind some related startups. This week I want to talk about a recent announcement from AWS that sheds even more light on what the future of serverless holds and why it will become the dominant force in cloud computing.

Let’s get started!

Why incremental changes can actually mean very big things… 🐘

Last week AWS announced that Lambda functions can now run up to 15 minutes. They also introduced the new “Applications” menu in the Lambda Console. On their own, these may seem like small improvements to AWS’s Lambda compute product. However, in my new post, What 15 Minute Lambda Functions Tells Us About the Future of Serverless, I argue that these announcement actually tell us a heck of lot more.

There are many arguments against serverless (some of them are even valid 😉). But these improvements give us insights into the ongoing battle to create a new kind of near limitless compute layer, one that could handle just about any use case. These are very exciting times.

When you realize that serverless might be more about business value… 📈

Every once in a while, someone will articulate something in a way that pretty much anyone (even company executives 😬) can understand. The Business Case For Serverless is a brilliant piece by Forrest Brazeal that should be sent to every C-level executive at your company. This piece covers everything from the total cost of ownership fallacy to the weak vendor lock-in argument. I also love how it doesn’t introduce serverless as an “all-or-nothing proposition”, but instead, can be adopted over time by using something like the Strangler Pattern.

When you’re looking for some serverless brain candy… 🍭

James Beswick’s (@jbesw) new post, Postcards from Lambda @ the Edge, is an exciting (and potentially sobering) look into the magical space that is Lambda@Edge. With all the recent buzz about Cloudflare Workers, it’s been easy to forget that AWS has had this for quite some time. The term “multi-region” is generally enough to induce panic attacks in many a developer. Lambda@Edge might be a good first foray into this uncharted (and mostly undocumented) territory. 👨‍🚀

Serverless DevOps: Why Serverless For Operations People is another great read where Tom McLaughlin thoroughly discuses how serverless makes teams more efficient by handing off operations to the cloud provider. Not only does this reduce operational costs, but it gives your team more time to solve business problems (you know, the things that actually make you money). 🤑

And Yan Cui is at it again, detailing some Common Node8 mistakes in Lambda. Node8 means access to async/await, but there are several ways in which we can (and most likely will) misuse this powerful new feature. This post is a must read for Node.js developers using serverless. 👩‍💻

Serverless Star of the Week ⭐️

There is a very long list of people that are doing #ServerlessGood and contributing to the Serverless community. These people deserve recognition for their efforts. So each week, I will mention someone whose recent contribution really stood out to me. I love meeting new people, so if you know someone who deserves recognition, please let me know.

This week’s star is Forrest Brazeal (@forrestbrazeal). I already mentioned Forrest’s amazing piece, The Business Case For Serverless, but Forrest is also a man on a (serverless) mission. As a Senior Cloud Architect at Trek10, Inc., an AWS consulting firm, Forrest works with real-world customers solving real-world serverless problems. But rather than keeping all that learning to himself, he is constantly sharing it with the community. Whether it be through his talks at conferences, his “Serverless Superheroes” blog series, his regular hosting of the Think FaaS podcast, or his FaaS and Furious cartoons, Forrest is doing great work to educate the masses about the benefits of serverless. And he’s also an AWS Serverless Hero.

Final Thoughts 🤔

For the last several years I have seen the same list of arguments against serverless: vendor lock-in, steep learning curve, not suitable for long running tasks, cold starts, poor logging, lack of observability, hard to test locally, response latency, etc.

Many of the above are valid concerns, but most are merely due to the fact that serverless is still in its infancy as a compute platform. Technical challenges aside, if we look beyond arbitrary limits, recycled cloud-based virtual machine arguments, and the semantics we use to describe it, we’ll see that something entirely new is being invented here. Serverless (or service-full) technology will continue to mature, and soon there will be little left to argue about.

How was this issue of Off-by-none? Feedback and suggestions are always appreciated. Contact me via Twitter, LinkedIn, Facebook, or email and let me know your thoughts, criticisms, and how you’d like to contribute to Off-by-none. Your input helps me make this newsletter better each week.

Go build some great serverless apps. Hope to see you all next time!

Best,
Jeremy