Good Morning!

It broke last week that AWS CMO Rachel Thornton is leaving Amazon and moving to Europe. I’ve more or less made fun of her entire org’s efforts for the last six years, but taking the Snark Pants off for a minute, AWS Marketing has built itself into something impressive and lasting. Her departure is very much a loss for AWS.

Also, a clarification for last week’s issue: when I said that someone should have been fired at IBM for buying Red Hat, my specific point wasn’t clearly stated: Red Hat was awesome, and I really fail to see any kind of evidence that it’s improved IBM’s culture any. It just reeks of a squandered opportunity.


From the Community

Data management is rapidly changing due to new technologies that focus on augmentation and automation. The Gartner® Hype Cycle for Data Management, 2022 will help you make decisions about the latest data management trends. Get your free copy of this 2022 Gartner Hype Cycle for Data Management, courtesy of ChaosSearch. Use this Hype Cycle to identify promising technologies and practices, plus decide when it is appropriate to evaluate them for adoption. While you are there, check out how companies like Blackboard, Armor and more are leveraging ChaosSearch to leapfrog their competition by massively scaling their access to critical business analytics.

This article on Working Backwards is one of the most insightful things I’ve read as far as shining a light into the mysteries of working at Amazon.

A specific cfn-init pitfall awaits you on the other side of this link.

Bloomberg’s Brad Stone has a fascinating piece talking about why Amazon’s acquiring things like One Medical and iRobot, companies that nobody externally wants to see them ruin.

Our Cloud Economics article about S3 Intelligent-Tiering: What It Takes To Actually Break Even has gotten some great feedback so far; check it out if you haven’t yet.

"The cloud providers should build out nuclear power" is certainly an interesting take.


Product Security at DigitalOcean helps solve large challenges while reducing the burden of security on dev teams, whether they’re building serverless function isolation or customer IAM. They believe security should make safe development easy. They’re looking for Senior Product Security Engineers who can collaborate with internal developers to design secure architecture and construct secure-by-default guardrails that empower engineers to make informed security decisions.


Last Week In AWS: Cadence Is Culture: Why Amazonians Need to Overload Us at re:Invent

Last Week In AWS: Twilio’s Insecure Text Message Issue

Last Week In AWS: Very Tired Lambda Pricing

Screaming in the Cloud: Brand Relationships and Content Creation with Jon Myer

Screaming in the Cloud: Creating Conversations on TikTok with Alex Su

Choice Cuts

Cut what you’re paying for cloud GPU instances by more than 73% by choosing Lambda Cloud. Access 1x NVIDIA A6000 (48GB) instances for $0.80/hour and 1x NVIDIA A100 (40GB) for $1.10/hour, compared to $4.10/hour for equivalent instances from AWS –kind of. In fact, AWS only offers one-size-fits-rich-kid instances starting at $32/hour to kick the tires. Imagine a world where the pricing was more accessible, transparent, and has no hidden fees. Sign up is free with no commitments.

Amazon Aurora Serverless v1 now supports PostgreSQL 11 and In-Place upgrade from PostgreSQL 10 – Ah hell they’re still adding features to v1, which means it hasn’t gotten the "Classic" treatment of being largely put out to pasture.

Amazon CloudWatch Evidently adds support for creating target customer segments for feature launches and experiments – Did someone on the CloudWatch team think about how this service name would make every headline mentioning it come across? Evidently not!

Amazon EBS adds the ability to take crash-consistent snapshots of a subset of EBS volumes attached to an Amazon EC2 instance – This is handy for those wild folks who attach 12 EBS volumes to a single instance but only want to have snapshots of some of them.

Amazon S3 adds a new policy condition key to require or restrict server-side encryption with customer-provided keys (SSE-C) – This one is for the compliance folks out there, who should be almost dancing in the streets over this.

AWS Console Mobile Application adds support for Cost Explorer service – This makes the AWS Console Mobile Application now the second most expensive app on your phone, after Robinhood. If only setting up multiple AWS profiles in the mobile app was less painful than passing a kidney stone…

New – AWS Private 5G – Build Your Own Private Mobile Network – Why would I want to do that? Now then: I’m amused by the lack of naming sensibility that went into this service. First, it doesn’t yet support 5G, so its name is at the moment an overambitious marketing promise. Next, when 6G comes out, what’s AWS gonna do? Obviously, they’ll launch a new version and call it "AWS Private 5G v2."

Building AWS Lambda governance and guardrails – Be forewarned: the first time your governance or guardrail makes Lambda harder for a developer to use, they’re going right back to servers or containers for the next six years. Tread lightly.

Estimating cost for Amazon SQS message processing using AWS Lambda – The real way that everyone estimates cost for AWS things is to run it for a bit, then do some simple arithmetic. WARNING: This can end poorly if you don’t check the pricing pages first.

Would you like to learn how to build inherently secure applications without jumping through time-consuming security hoops? Join our Live Hack Series on AWS where we’ll demonstrate how a malicious actor might exploit some of the most common vulnerabilities across several application components, and what you can do to apply automated security controls across the SDLC to quickly find and remediate those risks!

Introducing the new AWS Serverless Snippets Collection – So this is hosted on Serverlessland, which is basically an Amazon Basics version of Fine, I’m with you so far–but what about the Serverless App Repo? Nobody uses that thing anymore, including (apparently) AWS folks.

Amazon EKS now supports Kubernetes 1.23 – Why on earth would this not be branded as "Kubernetes 1-2-3?" Oh, right, because something that simple is already called ECS instead.

IPv6 addressing with Amazon RDS – A thorough dive into the nuts and bolts of what you’ll want to consider if you’re using IPv6 for RDS, which you probably aren’t yet.

Use Amazon RDS Proxy to provide access to RDS databases across AWS accounts – Credit where due, this one doesn’t feature traffic traversing the public internet.

AWS named for the first time ever as a Challenger in 2022 Gartner Magic Quadrant for Application Performance Monitoring and Observability – Having spent some time last week trying to get the AWS Distro for OpenTelemetry working in a relatively trivial Lambda environment, I’m kinda surprised Gartner rated them even this highly.

Announcing CDK for Terraform on AWS – I wonder what it would take to get my existing CDK code ported over to this thing. Probably more work than I want to invest, honestly.

Dashboards as code: A new approach to visualizing AWS APIs – When I attempt to visualize AWS APIs, I mostly just see a wall of red as my blood pressure spikes.


Presumably for legal reasons, Retool can’t describe what they do as “Visual Basic for internal apps,” but that’s how I think of them. I’ve been a long-time customer, and they remain the only low-code tool I’ve encountered that actually lives up to the hype. To learn more visit

When AWS breaks, its status page tells you it’s fine; the real story is on Twitter. This list of advanced Twitter search operators is depressingly useful to finding the answers to arcane questions about all manner of things–including the cloud.

… and that’s what happened Last Week in AWS.

Newsletter Footer

Sign up for Last Week in AWS

Stay up to date on the latest AWS news, opinions, and tools, all lovingly sprinkled with a bit of snark.

"*" indicates required fields

This field is for validation purposes and should be left unchanged.
Sponsor Icon Footer

Sponsor a Newsletter Issue

Reach over 30,000 discerning engineers, managers, and enthusiasts who actually care about the state of Amazon’s cloud ecosystems.