Off-by-none: Issue #185

April 26, 2022

Only have a few minutes? Check out this week's MOST POPULAR links as chosen by our email subscribers.

(Sort of) Serverless Aurora v2 😢

Welcome to Issue #185 of Off-by-none! This issue is sponsored by our friends at Lumigo.

In our previous issue, we were introduced to Function URLs, had a chat with Werner Vogels, and tried to stamp out serverless FUD. This week, Aurora Serverless v2 misses the serverless mark, Deno powers the edge, and we can now “Compose” serverless applications. Plus, we have lots of awesome content from the serverless community.

Before we begin, make sure you check out Lumigo | AWS Lambda Monitoring Platform | Get set up in minutes Sponsored

Serverless News & Announcements 📣

The biggest (and perhaps most anticipated) announcement these past two weeks was that Amazon Aurora Serverless v2 became generally available. Marcia Villalba wrote an excellent post explaining how it instantly scales to handle demanding workloads, TechCrunch mentioned it, and analysts saw it as another piece in AWS’s push for modern data strategies. I, however, was both underwhelmed and disappointed, and if you’re a serverless enthusiast, you probably will be too. 🙁

Also announced at the latest AWS Summit was that Amazon SageMaker Serverless Inference is generally available. This is actually very cool, and Antje Barth’s official blog post explains it quite well. Plus, George Anadiotis explains how it illustrates Amazon’s philosophy for ML workloads.

In other AWS news, AWS Amplify Studio is now generally available, and Amazon DevOps Guru now provides Proactive Insights to flag issues early on AWS Serverless Applications (which Marcia Villalba explains in more detail).

Some other interesting cloud news includes Netlify’s new Serverless Compute powered by Deno, Rookout added a serverless debug experience, and New Relic added Infinite Tracing for Browser, Mobile, and Serverless Applications.

Cloudflare announced their Green Compute open beta which only runs scheduled tasks from “green” data centers, RisingWave is now Open Source, and a new framework named Stacktape was introduced.

Google (predictably) donated the Istio service mesh to the Cloud Native Computing Foundation and Clumio announced Clumio Protect for Amazon DynamoDB, which I’m having a hard time trying to understand the value prop for.

Julian Wood reminded us that it’s Lambda planning OP1 time again, so get your requests in ASAP.

And finally, Serverless, Inc. introduced multi-service deployments via Serverless Framework Compose. This is an absolutely amazing feature that solves a lot of deployment issues. Yes, I do work at Serverless, Inc., but I’d be saying this regardless. This is just awesome work by the Serverless Framework team!

Serverless Concepts 🏗

Harinder Seera explains some strategies for making sure SQS doesn’t leave a hole in your pocket, Pablo Iorio outlines some serverless integration patterns, and Dustin Goodman gives you some tips for debugging Node serverless functions on AWS Lambda.

Daniele Frasca added three more posts to his Multi-Region Road series including one on Amazon API Gateway, another on Amazon DynamoDB, and the final one talking about our old friend, caching.

Tim Davis shared a post about solving for deprecating runtimes in your serverless applications, which suggests several scenarios in which it might be easier to “replatform” your Lambda functions. Maybe in extreme scenarios this makes sense (like Python 2.7 to 3), but I’m sure there are custom runtimes out there that you could use to avoid something so drastic.

There’s a post that compares MongoDB vs DynamoDB and asks “Which NoSQL database is right for you?” Overall it a pretty good overview, although the comments provide a few necessary corrections. There’s also this post about accelerating serverless development with AWS SAM Accelerate. Testing in the cloud should just be the way. 😉

And Tomasz Łakomy provides a list of the 8 best practices for optimizing Lambda functions.

Serverless Tutorials 👷‍♀️

Here are a few tutorials to keep you busy (and learning).

Wesley Cheek shows you how to implement AWS Lambda Function URLs with AWS CDK.

Matt Martz thinks he invented CDK Level 4 Constructs, and shows you how to use them.

This tutorial shows you an easy way to set your Serverless Framework stage name using an environment variable from GitHub Actions.

And Doug Moscrop shares how to use Serverless Cloud with SvelteKit.

Serverless Reads 🤓

There were a ton of really interesting reads these last two weeks, so here’s a sampling of the ones that got me thinking.

Let’s start with Tom Naylor’s The Thing About Serverless. I love his conclusion that “the advantages are so overwhelming and obvious at this point,” that he has, “a hard time imagining doing it any other way.” But he also focuses quite a bit on reducing costs, or more accurately, Total Cost of Ownership. Paul Swail has a piece that talks about the problem with TCO, which focuses on issues convincing decision makers to play the long game.

Joe McKendrick’s No Ops like ‘NoOps’ revisited: is the vision a reality yet? touches upon allowing teams to focus on “value-added activities.” But we’re seeing some companies building out “serverless enablement teams” that have ops people dedicated to enabling better and more automated software delivery for their developers. #BetterOps is a more promising future IMO.

Sustainability is becoming a more popular topic (and probably not just because of Earth Day). Bonnie McClure shared a collection of AWS posts on designing sustainable cloud architectures. There was also this Ask an Expert post on sustainability that seems to suggest that AWS isn’t ready to make their customers feel bad about not going serverless. Meet customers where they are I guess.

There’s another kind of “sustainability” that’s also important, which Sheen Brisals explains in his piece on how to practice sustainable development processes in serverless.

Salvian Reynaldi has a series on embracing serverless that demonstrates that the struggle is real, and perhaps the solution isn’t always obvious. And maybe this is because many engineers don’t understand serverless, as Anna Geller suggests.

Vincenzo Marcella tells a tale of programmatic disasters with scheduled Lambda functions. This type of problem underscores the need for single purpose functions, preferably orchestrated with Step Functions with failure steps that implement rollbacks using a SAGA pattern. If it sounds complicated, that’s because it sort of is. But it’s also highly reliable.

And speaking of disasters, Allen Helton asks if disaster recovery is worth it in serverless applications. You probably first need to understand what disaster recovery actually means, which Allen explains.

Shaked Lokits’ post on 15-Minutes-Driven-Development, or why you should use a server, highlights the mindset that is holding developers back and limiting their ability to realize the full value of serverless. It’s the old way of single-thread thinking that misses out on the benefits of parallelism, redundancy, scalability, and resiliency.

If you’re looking for more to read, check out the 20 best serverless blogs to keep you in the loop. And you can also check out Who’s Who in Cloud? Somehow, I made the list.

Podcasts, Videos, and more 🎧

There were two great episodes of Serverless Chats including Episode #133: Moving to Serverless Safely with Jeff Williams and Episode #134: Serverless Community Building with Farrah Campbell.

Marcia Villalba released two great videos. One highlighting what’s new in AWS CDK v2, and another on choosing the right storage solution for AWS Lambda (AWS S3, AWS EFS, Lambda layers, or temp storage).

iLyas BK shared a video about “Denonia” the first publicly-known “virus to target AWS Lambda”. This video does a good job explaining how it works, but since (as we all should know by now) it doesn’t take advantage of any actual exploits, he simply gives the advice to protect your account credentials.

And here are two Serverless Office Hours for you that explain Amazon EventBridge Global Endpoints and AWS Lambda Function URLs.

New from AWS 🆕

Besides the aforementioned announcements, AWS has been quite busy.

• Introducing 35 new resource types in the CloudFormation Registry
• Amazon Textract launches new Queries feature within Analyze Document API
• AWS Step Functions expands support for over 20 new AWS SDK integrations
• New data source connectors generally available for Amazon Athena
• AWS Amplify Studio Figma-to-React code capabilities are now generally available
• AWS App Runner launches AWS X-Ray support
• Amazon Neptune now offers a free trial
• Announcing consolidated view of Lambda Insights via Application Insights
• AWS Fargate now delivers faster scaling of applications

Thoughts from Twitter 🐦

I’m seeing in person meetings already and the big challenge becomes – how do you not make remote workers isolated? If 5 people who get in a room together feel fantastic but leave out everyone who wasn’t there, will remote first competition not have the best remote talent? ~ Matt Coulter
Matt is commenting on a larger thread that has more observations of what a post-pandemic workforce will look like. Remote work was a thing long before Covid, but it’ll be interesting to see the cultural shifts from its massive impact on knowledge workers.

I cannot believe AWS is forcing everyone to go to the new DynamoDB console, considering it’s unusable (vs the old) for editing records. Wondering whether @rafalwilinski is somehow behind this because we’ll all be using Dynobase by mid June because of this. ~ Joe Emison
I spoke with a PM about the new DynamoDB console months ago and expressed my concerns. Maybe they’ll listen to Joe? 🤷‍♂️

Upcoming Serverless Events 🗓

If you have an event, webinar, etc. that you’d like me to mention, please email me.

May 3–5, 2022 – Reactathon 2022 & Serverless in the Park

June 22, 2022 – ServerlessDays Paris 2022

June 24, 2022 – ServerlessDays New York 2022

Serverless Star of the Week ⭐️

There is a very long list of people who are doing #ServerlessGood and contributing to the Serverless community. These people deserve recognition for their efforts. So each week, I will mention someone whose recent contribution really stood out to me. I love meeting new people, so if you know someone who deserves recognition, please nominate them.

This week’s star is Eder Rueda Fernandes (@devawsbr). Eder is a Solutions Architect at GFT, with a focus on Cloud, Open APIs and Open Banking. He’s also responsible for Open API Framework, an accelerator to your Open API journey produced by GFT in partnership with AWS. Additionally, Eder has a YouTube channel where he provides videos with the purpose of making AWS simple (for any and all Portuguese speakers out there). Thank you, Eder, for your commitment to helping others in the community!

Final Thoughts 🤔

It’s good to be back, but wow was that a lot of content to catch up on. So much to learn and stay up to date with, but always amazing to see how everything is progressing.

And… don’t forget to sign up for email updates on my upcoming DynamoDB modeling course.

Cheers,
Jeremy

I hope you enjoyed this newsletter. We’re always looking for ideas and feedback to make it better and more inclusive, so please feel free to reach out to me via Twitter, LinkedIn, Facebook, or email.