Our serverless superhero this week is Anna Spysz, front-end engineer at AWS and amazing public speaker and serverless advocate. Anna works on the Infrastructure Composer (formerly App Composer) team, helping to fundamentally change the way we write infrastructure as code. She has a true grasp for developer experience, which holds a near and dear place in my heart. Thank you for everything you do, Anna!
AWS Lambda turned 10 years and Werner Vogels shared with us something that is usually never shared… the original Lambda PR/FAQ. The document that started it all and has been followed by many of the services we now use as part of the serverless ecosystem. It’s incredible to see the vision they had back then and see the evolution it has gone through over the past 10 years. What they achieved with this innovation has allowed us to build so many things without having to pay special attention to all the managed infrastructure behind it. I’m excited to see how the cloud evolves over the next 10 years, it’s probably going to be a whole different landscape.
It’s probably not news for you but caching is hard, and if done wrong it can cause some serious headaches. I love to see other peoples approach to this problem and Luc van Donkersgoed shows us exactly how he managed to add caching to make the AWS News Feed super fast in preparation for re:Invent. This is a great read with very applicable insights.
GPU workloads are expensive. Wouldn’t it be nice to dynamically route workloads to the whichever cloud that has the lowest cost and highest GPU availability? Good news - you can! Beta9 is an open-source platform for running AI workloads across any cloud environment. Engineering teams can deploy inference endpoints and run training jobs with autoscaling, scale-to-zero, and automatic clean up of idle resources. Beta9 can run workloads on any infra, including bare-metal servers, VMs, as well as all the major cloud providers. Check out and star the repo today! Sponsored
AWS recently released a feature to be able to encrypt your Lambda functions artifacts using your own Customer Managed Keys. Rahul Ladumor gives us a great walkthrough on his latest post and it even includes a CloudFormation example.
Cloudflare seems to be making a lot of appearances everywhere, don’t get me wrong they are doing great great stuff. On his latest post Kirk Kirkconnel talks about the power of combining Fauna and Cloudflare workers. I really like the idea of including native integrations like these to make it even easier for us to create applications with less friction.
It sucks saying this, but most of us typically leave security until the very end. But it doesn’t have to be that way, there are now services that really bring down the barrier for us to secure our applications. AWS WAF is one of them. In his post Ran Isenberg goes through AWS WAF essentials, the benefits and how you can use it to add an extra layer of protection to your applications.
Another great post by Yan Cui this week where he talks about when to use light events vs. rich events in event-driven architectures. Learn about the pros and cons of each so you can use them appropriately.
Documentation sucks! There I said it, the worst thing is I’m not the only one that thinks this. What is even worse is that it is very important to have good documentation to be able to keep moving an application forward. So how do you make sure your application has documentation and it doesn’t go stale?. David Boyne shows us how you can keep updated documentation while still being happy. I really need to go try docasaurus now.
Gotta enjoy the little wins when you make them.
— David Boyne 🚀 (@boyney123) November 15, 2024
Awhile ago I invested time to automate SDK documentation through code docs (with docusaurus), which is now paying off 🚀
1. Make changes to SDK
2. Update docs in SDK
3. Auto released on the website
Love docusaurus and it's custom… pic.twitter.com/Imla8LMbMj
pre:Invent is on full swing right now and you can feel it with all of these great announcements. I really hope these keep on coming through re:Invent as well.
There is now a way to centrally manage root access in AWS IAM, this is really helpful to maintain higher security standards and simplifying the management of privileged users across your organization.
AWS Step Functions can now export workflows as plain AWS CloudFormation or AWS SAM from the console (sorry CDK users). I typically start creating workflows from the console and being now able to get the full IaC from it is going to save a some time.
A couple of DynamoDB goodies. First one is that pricing for on-demand throughput has been reduced by 50% and global tables by 67%. And second, DynamoDB introduces warm throughput for tables and indexes, giving us the ability to pre-warm our tables in advance for peak traffic events, the best thing about this is that it is automatically there for you for no extra cost.
I am super excited about this one, we can now see a timeline view for AWS CloudFormation deployments. This really helps to get an idea of which resources are taking up most of your deployment times. With this information you can potentially split your slower resources into separate stacks to make your day to day deployments faster.
You can really feel the pre:Invent vibes with these announcements. If you want to keep track of all of these as they come out go to AWS News Feed where there is a special section where all of these are coming in.
What a week! We celebrated 10 years of Lambda, there was so much good content and so many announcement. Thanks for letting me be part of your day. Have an awesome week everybody!
Until next time!
Andres
Thank you for subscribing!
View past issues.