Good morning!

Welcome to issue number 97 of Last Week in AWS.

Last week I went to Seattle, caught up with a bunch of people who are way smarter than I am, and made fun of their work at a comedy show. There’s another one of those next month, tentatively on March 20th. If you’re in Seattle, watch this space, or my ridiculous twitter feed.

Other than that, absolutely nothing of interest happened in Amazon news last week. Not one single thing! How’s the east coast doing these days? Weather good? How’re the Yankees doing this year? Good chat!

This issue is sponsored by Scalyr. Learn new Kubernetes deployment and development processes with Scalyr during an online seminar on Tue 2/26 at 10:30am PT. Steven Czerwinski (Scalyr CTO and co-founder) and Dave McAllister (Scalyr Community Guy) will be explaining how new infrastructures need new approaches for deployment and reliability. They’ll be showing you how to achieve performance at scale, and will also share conceptual frameworks to apply to your work. Sign up here. Thanks to Scalyr for their continued support.

Community Contributions

If you’re unclear on what AWS Lambda actually is, don’t be ashamed–be enlightened, with a quick read-through of AWS Lambda in Plain English.

The Wrong Way to Do Things Slapfight Championship quarterfinals heat up with a throw-down comparison between Ansible and AWS Systems Manager’s RunCommand.

The Cognito / Amplify teams would like your help untangling their labyrinthine authentication interplay.

I’m a fan of cost analyses, and this V100 server on-prem vs AWS p3 instance cost comparison is no different. While there’s a story around care and feeding of hardware that you get to avoid with AWS, this comparison includes it. It’s… pretty clear that the p3 instance family has some cost cutting to do if it wants to remain competitive for steady-state workloads.

A dive into how Marqueta handles pre-signing of S3 URLs, something that people keep forgetting is there and needing to be reminded about–or is that just me?

AWS legend James Hamilton gives a great rundown as to just what the hell Nitro is.

Jerry Hargrove has updated his Periodic Table of Amazon Web Services.

AWS has an increasing challenge when it comes to narrative; both Microsoft and Google are pitching their clouds with a simple message: we won’t compete with you. You can get into a nuanced debate here–but only if you’re invited to the conversation.

Yet more AWS S3 security best practices, this time focusing on the various ways to securely control access.

A story about how Intercom uses DynamoDB Streams and some code to visualize changes in frequently updated DynamoDB objects.

Last week it became clear that Steve Ballmer was an incredibly successful deep-cover agent bent on destroying Microsoft, as the NBA team he owns selected AWS as their official cloud provider.

I spoke with Elliott Murphy of KindlyOps about compliance, data retention, and working with regulated companies in AWS. Tune in for Episode 48 of Screaming in the Cloud: Nobody Gets Rid of Anything, Including Data.

Choice Cuts From the AWS Blog

Amazon EFS Introduces Lower Cost Storage Class – It’s now cheaper to use EFS; a single checkbox separates you from 85% cheaper storage for files you don’t use frequently. There’s (almost) no fiscal downside here.

Amplify Framework Adds Support for Multiple Environments, Custom Resolvers, Larger Data Models, and IAM Roles Including MFA – Amplify gets more capable, and (since it’s named Amplify) INCREASINGLY LOUDER!

Deploy IBM Cloud Private on AWS with New Quick Start – Oh, a QuickStart snuck its way in somehow. AWS has a quickstart out that tells us how to install IBM Cloud on AWS and now I’ve gone cross-eyed trying to make sense of what this is.

Introducing Five New Amazon EC2 Bare Metal Instances – The titanic game of Battleship between two AWS VPs via instance family names continues as one of them evidently sunk a destroyer.

Introducing AWS Solutions: Expert architectures on demand – “AWS Solutions” is a bunch of implementations of existing architectures to solve problems. It’s well named, its solutions are well built and exceedingly well documented, and so I can only assume that this was released behind somebody’s back. Keep an eye on this; it’s a lot better than AWS’s existing “QuickStarts” that keep trying to sneak their way into this newsletter.

This week’s issue is sponsored in part by GoCD. This week, they have an article discussing lead time in the context of CI/CD, and why you might care about it. Figuring out how long it takes to go from development to production is awfully important when answering business stakeholder questions honestly; this one’s worth a read. Thanks to GoCD for their continued support of this newsletter.


A cloud-agnostic way to compare and contrast instance types between providers, cloudinfo is worth a glance if you’re trying to comparison-shop.

Leftovers is a Go library (with accompanying CLI) for cleaning up orphaned IaaS resources across a variety of providers. This is the kind of thing a kind cloud provider would do for you automatically.

…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.