Our serverless superhero this week is Praneeta Prakash, Product Manager at AWS. Praneeta is responsible for many of the amazing features we see in the SAM CLI today. Without tools like the SAM CLI, the serverless world would be a very different landscape. Probably one of the friendliest people I know at AWS, Praneeta is always willing to receive feedback and look for clever ways to solve pain points. Thank you Praneeta, for enabling the community to get to the cloud fast and easy!
When you don’t know what your traffic will be in your app, it is often recommended to configure DynamoDB to use on-demand capacity. This allows it to scale while you discover your workload. But what do you do after you’ve identified the amount of traffic? Pascal Martin tells us how to calculate DynamoDB RCU and WCU capacity on dynamic workloads. He walks through the theory behind provisioned capacity and why making reservations can save you big money in the long run.
In the spirit of optimization, Ali Haydar tells us how to optimize your Lambda performance with NodeJS top-level await. If you need to perform tasks like looking up a secret, he shows us how to do so outside of the Lambda handler, allowing for that operation to be done during initialization instead of runtime. It’s a great way to reduce latency in your functions.
We all saw App Composer released at re:Invent - the exciting new service that allows you to visually build your cloud resources and save them as a SAM template. Jimmy takes us end to end in the development process using App Composer. He walks us through building a serverless AWS announcements email service using App Composer to generate the IaC and finishes it out by implementing the function code and deploying with SAM. It really shows the speed of going from idea to working in the cloud using the tools we have available today.
Databases collectively are often a top-three line item on large cloud customers’ bills, and even though backend caching is one of the best ways to improve performance, stability, and cost efficiency, it’s not always implemented. Read this blog to learn how to optimize your database while cutting down costs by almost 50% by adding a serverless cache. Sponsored
Observability is often underestimated in both importance and difficulty to implement in serverless apps. Developer Steve sheds some light on why that is with his post on the hidden costs of serverless observability. He talks about why it’s so difficult, what you need to monitor, and answers the build vs buy question when it comes to serverless observability.
Pablo Bermejo gave us his take on what’s next after serverless. His message revolves around the serverless development experience and making it better with things like Infrastructure from Code (IfC). It’s worth a read if you want a glimpse into the future.
Did you see AWS released a serverless digital learning badge recently? Daniel Aniszkiewicz shares his experience with it and gives us his first impressions. He tells us what to expect and if the badge is worth the effort.
A couple of big updates came to the NoSQL Workbench last week. It now comes with DynamoDB local allowing you to quickly get started modeling and testing on your local machine. It also now supports creating data models directly from sample data model templates which gives you an accelerator to modeling your data.
We’ve gone back and forth the past few months on the whole what is serverless game. Does it simply mean “as a consumer I don’t manage servers”? Or is there more to it? This is a topic that has been gaining momentum in the community lately and has started building polarizing opinions. So I, Allen Helton, decided to step right into the fire and throw out my two cents with my post stating I don’t know what serverless is anymore. It’s a bit tongue-in-cheek, but attempts to make some differentiations with what serverless actually means.
As is tradition when a cool new EventBridge feature comes out, David Boyne is all over it with practical examples and working code. I found his example of DynamoDB Streams to EventBridge Bus particularly useful.
Exploring a pipes pattern, that allows you to listen to DDB events, split them out into separate pipes using filters and raising events directly onto EventBus.— David Boyne 🚀 (@boyney123) December 7, 2022
In theory you could raise domain events, without the need for Lambda / Custom code.
Listen, filter, and integrate pic.twitter.com/I8ZV19Xw8t
In unrelated news to serverless, last week I published my 50th blog post for the year! If you don’t follow the blog, there are lots of good articles on serverless all in one place - I definitely recommend checking it out!
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.