Another week gone by, and we’re well into the doldrums of Winter. Stay safe / warm / frugal…
From the Community
Get a handle on cloud sprawl! Empower teams to deliver better insights, faster. CloudQuery is a high-performance data integration platform for security and infrastructure teams. Sync your configuration data from any cloud or SaaS application to any destination/database to build your asset-inventory, CSPMs with standard data and visualization tools.
“The Commonwealth of Amazon” over at Public Seminar talks about a growing problem: as Amazon continues to grow, it invariably captures the political process as well. Once again I wish AWS would be spun off into its own entity.
The 2022 retrospective of Follow the CAPEX mentions that Amazon is still investing 35% of AWS revenues back into cloud infrastructure. That’s just a monstrous amount of money… Say what you will about the company, they’re not cheaping out on providing capacity.
Over in Azure Land apparently some customers have been told that real-time tracking and notifications of customers’ Azure spend is broken until mid-March, which is just WILD. "Yeah, you can’t see when you overrun your budget until another month from now" is indicative of something being deeply, deeply wrong with the revenue generating bits deep within the bowels of the cloud.
Former AWS employee Zack Glick writes about Using Jupyter Notebooks in Google Colab and publishing to Google Sheets and I love it. This is effectively what Amazon SageMaker wishes it was and could be if it focused on being approachable and user friendly rather than mercilessly expensive.
Last Week In AWS: Telling Customers What They Want To Hear
Last Week In AWS: The AWS Community Isn’t for Amazonians
Last Week In AWS: Wait did you say “Drone Manufacturer?!”
Screaming in the Cloud: The 4D Approach to Cloud Sustainability with Catharine Strauss
Screaming in the Cloud: The Ever-Growing Ecosystem of Postgres with Álvaro Hernandez
DevOps practices have revolutionized how apps and infrastructure are managed, but access hasn’t kept up. Shared secrets like passwords and keys – the #1 source of data breaches – are the norm. Teleport replaces shared secrets like passwords, keys, tokens, and even browser cookies with true identity, removing risk while letting engineers go fast. Check out Teleport today.
Amazon Chime SDK now offers a Windows client library – This makes sense. People using Windows are already miserable, may as well give them a decent messaging backend. I’m serious! The Chime SDK is awesome; it’s the app that hates people.
Amazon CloudWatch now supports high resolution metric extraction from structured logs – All in service of CloudWatch’s low resolution wallet extraction, presumably. I love CloudWatch, hate the lack of visibility into the specific cost drivers so far.
AWS SAM CLI introduces ‘sam list’ command to inspect AWS SAM resources – I see “list” won the final round coin flip, bearing out “describe,” “get-status,” “ls,” “ps,” and for some reason “systems manager listing manager.”
Get cost estimates faster with AWS Pricing Calculator bulk import – I’d like it if I could get order-of-magnitude billing information without waiting a day or two first; what kind of bulk import do I need to build to make that happen?
New – Visualize Your VPC Resources from Amazon VPC Creation Experience – Okay I like this a lot. Finally I don’t have to click through six screens and keep track of VPC / Subnet IDs to figure out whether an S3 gateway endpoint is in place or not. That alone is worth however much effort it took to build this.
Introducing the AWS ProServe Hadoop Migration Delivery Kit TCO tool – There’s just no way that whoever named this thing isn’t being paid by the word.
Quick Restoration through Replacing the Root Volumes of Amazon EC2 instances – Huh; have OSes gotten to the point where reusing a root volume on a new instance doesn’t lead to misnamed devices / the system being forever wonky? Because that would be awesome if so.
Introducing the Amazon EKS Workshop – I’m picturing something akin to Santa’s Workshop immediately after one of the elves grabbed a sharpened candy cane and led a failed yet surprisingly bloody uprising against their ongoing exploitation.
Using GitHub Actions with Amazon CodeCatalyst – This is yet another example of how CodeCatalyst "gets it" in a way that most of the other service teams very clearly do not. Instead of shoving customers towards CodeBuild / CodePipeline / CodeDeploy (collective motto: "Bet you’ll use GitHub Actions next time, loser!") they’re meeting the industry where it is and making it straightforward to use GitHub Actions.
Consolidate and query AWS CloudTrail data across accounts and regions using AWS CloudTrail Lake – Oh hello; unified CloudTrail Lake? I am very much here for this. It’s hard to overstate just how great this service is.
Using Amazon CloudWatch metrics to monitor time to expiration for Reserved Instances – In a smarter world, this would already be exposed, but no; you get to build (and pay for) CloudWatch custom metrics to freaking see when your RIs expire. Genius.
MinIO object storage runs everywhere the cloud operating model runs – offering S3 compatible, cloud-native storage to enterprises that value simplicity, scale and performance in a software-defined, self-hosted solution. Learn more at www.min.io and be sure to tell them that Corey sent you.
ses-sidecar is a way of getting fixing the semi-annoying problem that SES’s SMTP service doesn’t work with temporary credentials. Welp, now it does!
former2 is worth another look; it generates CloudFormation / Terraform / Troposphere templates from your existing AWS resources.
… and that’s what happened Last Week in AWS.