Serverless Superhero
This week’s serverless superhero is Ala Shiban, founder and CEO of Klotho. Ala has been pioneering the infrastructure from code space and just recently released the closed beta of Klotho. He has lots of great ideas on how to bring serverless compute to the hands of many. Thank you Ala, for all your hard work and congratulations on the release!
Tutorials
We talk a lot about event driven architectures, but sometimes fail to mention specific patterns that make them work so well. Bobur Umurzokov helps us out by talking about CQRS, event sourcing, and serverless in his blog last week. This post doesn’t describe in detail how to implement with Lambda or EventBridge, but rather it talks about the moving parts of these complex patterns. A great way to expand your scope of understanding for these difficult concepts.
If you want specifics about building event-driven patterns in AWS, Marcia Villalba posted a great video on exactly that. She walks through building an example pub/sub project using the CDK. Her video demonstrates how to build this pattern using SNS rather than EventBridge.
Ever wonder why you’d use SNS over EventBridge? Lee Gilmore has another outstanding article where he talks about guaranteed event ordering. He describes in detail how to build an event system with SNS and EventBridge living side by side to enforce ordering when necessary. As usual, Lee provides a reference architecture written in the CDK for you to follow along.
I was recently trying to figure out how to log VTL errors from API Gateway integrations into CloudWatch. Luckily for me, Binh Bui published his post on AWS API Gateway access logs where he shows us how to enable, format, and query access and execution logs. Best part? He shows you how to enable this in the CDK, so you can take it and immediately start enabling logs in your apps.
A couple months ago AWS released an update that allows CloudFormation to publish EventBridge notifications, unlocking a lot of devops-y potential. Mahanth Jayadeva demonstrates how to build custom workflows for post provisioning management. He talks about what could happen when a stack fails to deploy or update and how to respond to it.
Spotlight
CDK is taking the cloud world by storm. The adoption rate keeps climbing, we’re seeing more and more reference architectures built using it, and many blog posts use it for quick launches. It’s a great tool that provides a level of abstraction we haven’t had access to before. But it’s not perfect. Matthew Bonig covers the bad parts of the CDK in his video where he discusses everything from semantic versioning, to determinism, to ownership, and much more. There is a ton covered in this video and is absolutely worth the watch whether or not you’re an active CDK user.
Read here how you should stop settling for almost serverless. You aren’t fully serverless until your cache is serverless. We have serverless functions, storage, and databases. But when it’s time to accelerate with a cache, the fully serverless dream falls apart. Not anymore! Sponsored
Tip of the Week
There tends to be confusion on what serverless is or isn’t. Anna Geller shares with us an incredibly thorough and easy to follow thread to help clarify for us. 👏
#Serverless is still misunderstood.
— Anna Geller (@anna__geller) September 18, 2022
It’s not about servers or lack of them.
It’s an operational construct 🧵
Last Words
Lately I’ve been taking a dive into infrastructure from code solutions to see where we’re at as an industry. I’ve formed some interesting initial opinions and am curious to see how some of these vendors play it out. There’s lots of potential here to build something revolutionary, but it will be tricky along the way.
If you’d like to make a recommendation for the serverless superhero or for an article you found especially useful, send me a message on Twitter, LinkedIn, or email.
Happy Coding!
Allen
