Serverless Superhero
Our serverless superhero this week is Alvin Johansson, software engineer at Elva and AWS community builder. Alvin recently created a game to see if you can recognize AWS service logos at a glance and offers some great insights in online serverless groups. Thank you for all your contributions to the community, Alvin!
Tutorials
Starting us off is Lee Gilmore with a fantastic post on how to schedule emails with EventBridge Scheduler and SES. This article walks you through a highly-scalable architecture that does exactly what you think - sets up one-time schedules to send an email. I find this to be incredibly practical and a great reason to start trying out this functionality of EventBridge if you aren’t already. Code samples are included in the article for you to try it yourself.
Cloud cost management was by far my least favorite responsibility as a Cloud Architect. Tracking usage, optimizing resources, and forecasting usage accurately to take advantage of usage discounts aren’t my strong suit - app building is. That’s where companies like Archera come in handy. Their platform not only shows where your cloud spend is going, but it also helps you plan smarter with their automated cloud cost optimization tools. They take the complexity out of managing your spend, which means everything in today’s world. Hand over the responsibilities of pricing negotiations, maximizing cloud credits, and finding the right discounts so you can get back to what you really love. Check out some of the neat things Archera can do for you. Sponsored
The promise of EventBridge pipes is to simplify how you connect AWS services. That said, I’ve found implementing them anything but simple. Every time I go to implement a pipe, I feel like I have to relearn everything again. That’s why I like Derek Murawsky’s blog from last week where he shares how to use them in practice. His post explains what they are and the different components that make them up, plus he shares a CDK construct you can use to get started in an instant. I like articles like this that help launch you and keep you going project after project.
Interesting Content
Two weeks ago we saw Elasticache and MemoryDB got some Valkey updates. In case you couldn’t quite tell how big of a deal that was, Marcin Sodkiewicz did the math for you in a quick review last week. No spoilers here, it’s a 3 minute read with lots of pictures, go check it out.
I found Seth Orell’s article on testing event-driven architectures both thought-provoking and bold/funny. Starting off with a “not testing is a good strategy” claim, he offers his opinion on why other common forms of testing might not be as good as you think. His actual recommendation is a great idea and has lots of room for growth and modification based on your use case.
James Eastham has been doing some wonderful work lately around high-performance applications. Notably, he’s been focusing on Rust and Cloudflare. His video last week expands on his series by getting into the details of high-performance Postgres. Of course, this is still on the Cloudflare and Rust train, but how could it not be if you’re trying to make something as fast as possible? James walks through everything from directly connecting to your DB to local development to troubleshooting and best practices.
Spotlight
I’m convinced David Behroozi is a mad scientist. The experiments he runs are not only super thorough, but also incredibly interesting. In his post last week, he shows us how he measures latency in his CloudFront functions. The execution is genuinely innovative and I love how it makes me think a little bit differently. He walks through the problem, solution, metric details, and costs of everything. Fantastic work and way to think out of the box, David!
Tip of the Week
Definitely not a tip, but the meme James Beswick shared last week made my day. It also begs the question - are you happy with the schedule at AWS re:Invent this year?
Quick summary of the AWS re:Invent session catalog...#aws #serverless #reinvent pic.twitter.com/LlkmhNjtz3
— jbesw (@jbesw) September 24, 2024
New Releases
This time of year tends to be the calm before the storm as far as AWS updates go. Service teams are getting their major feature releases ready for re:Invent and the few weeks around it. That said, there are a few things noteworthy from last week.
CloudWatch Evidently has been deprecated in favor of AppConfig. This is actually a good call, in my opinion.
AWS has also discontinued Kinesis Data Analytics for SQL starting January 27, 2026. There’s plenty of lead time on this one for migration, but the rub on this one is that the deprecation announcement is hidden in a migration blog post. Kudos to Jeremy Daly for spotting this one.
DynamoDB released a “UX enhancement” last week, allowing you to favorite your DynamoDB tables. Your starred tables will now show up under a new “favorites” icon. As far as I can tell, this is just an AWS console thing.
Lambda also received some UX enhancements, adding in CloudWatch Logs live tail and Metrics Insights to the Lambda console.
Last Words
If you haven’t seen it yet, the Believe in Serverless community is releasing daily production horror stories from real-world practitioners. These are great reminders that we all make mistakes and that we should use them to recover, learn, and improve.
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
This Week's Sponsor
