Our serverless superhero this week is Matt Carey, serverless developer at aleios and AWS Community Builder. Matt organizes the Serverless London meetup group, bringing serverless enablement to hundreds of people each year. He also is an ex-professional windsurfer (super cool!). Thank you Matt, for everything you do for the serverless community!
Serverless applications tend to be microservice-based. Each microservice is isolated with access to its own resources. But end users don’t need to know or care about how you isolate resources. They want a consistent, unified way to access data and make changes. So how do you consolidate APIs across microservices into a single, cohesive API? Lee Gilmore walks us through API Gateway federation to do just that. In typical Lee fashion, he gives us an incredibly detailed example with brilliant diagrams and easy-to-follow explanations.
I’m a big fan of using AI to help out with everyday tasks. So is Mohamed Labouardy. He published an article last week on how he built a gym app with ChatGPT to generate workouts. He uses Twilio to send the workout straight to your phone. It’s a nice solution that takes advantage of the tools we have available for free to make life a little easier.
We’re all in different places in our cloud journeys. Some of us have been deep into serverless for years, while others are just beginning to dip their toes in the water. For those who are starting to explore the idea of serverless, I (Allen Helton) wrote a blog as a reminder that you have non-engineering considerations to make when you go serverless. I talk about some of the troubles I had when I started and remind readers of a few of the trade-offs you make when you decide to incorporate serverless services.
Managing scale in serverless applications can be tricky. The infrastructure can sometimes scale too quickly and drown downstream services with an influx of events. There are a number of ways to mitigate this, with one of them being API Gateway usage plans. Moudud Abu wrote a great article talking about how you can implement a usage plan in API Gateway to throttle requests from API consumers. This not only helps your infrastructure stay stable but also prevents a single tenant from running away with all the cloud resources in a multi-tenant environment.
Andrea Falzetti published an article on lessons learned over 5 years and 100 functions in production where he shares some opinions he’s formed over the years. He has some good and bad realizations, along with some recommendations. While I don’t necessarily agree with all his conclusions, he brings up some astute observations from the production battlefield.
Ampt released configuration-less WebSocket support last week. It’s a huge step in the right direction for serverless applications that need real-time communications between a browser and a server.
The CDK received an update to validate IAM policies during synthesis time. This provides build-time security policy validation to prevent any accidental mishaps from making it to the cloud.
AWS App Runner added 7 new compute configurations. Included in this offering is a significantly reduced minimum configuration, dropping the lowest possible cost to run by almost 4x!
AWS released the Q1 Serverless In Case You Missed It. So much happened in AWS serverless in the first three months of the year, this article is gold!
I spoke with Boris Tane and Thomas Ankcorn on the Ready, Set, Cloud Podcast about developing fast without being reckless. We talk about horror stories, how they decide what to work on next, and decide if moving fast is for everyone.
Yan Cui had Dax Raad on the Real-world serverless podcast to talk about building the open-source community around SST. It’s a great listen with a lot of lessons hidden inside.
As we all know, it’s the year of observability. In the spirit of broadening our understanding, Marcin Sodkiewicz wrote an article telling us about AWS & OpenTelemetry. This article has everything you need to know about OTEL and he also describes how you can capture all the native AWS OTEL data yourself without implementing a third-party solution. It’s a great educational piece about a topic we often hear about but don’t dive into. Put your learning caps on!
Lars Jacobsson continues to make incredibly valuable improvements to his sam-patterns-cli
package. Two weeks ago he introduced ChatGPT into the mix to generate SAM templates for you. This past week, he added the ability to describe a SAM template and analyze them for security issues and offer recommendations.
💥New sam-patterns-cli command: 'describe'
— Lars (@lajacobsson) April 6, 2023
Point it at an #AWSSAM template, local or in github, and it'll use #ChatGPT to summarise it for you, point out any security issues and give suggestions on how to fix them.
As it's an Easter release, there's an easter egg🥚
GPT-4… pic.twitter.com/QQZAenertM
Are you interested in writing articles for Ready, Set, Cloud? I’m entertaining the idea of opening the doors to other authors and would like to gauge interest. Please let me know if you’d like to write along with me!
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
Thank you for subscribing!
View past issues.