It looks like we’re getting back to “normal,” finally. AWS is apparently in peer-review season, with compensation adjustments to match. If you work at AWS and like money, you’re going to be very, very interested in my blog post / email / podcast trio this Wednesday. Keep an eye peeled if you have some spare time while showing up in person for unionization votes.
From the Community
This issue is sponsored in part by my friends at ChaosSearch! As you know, log analytics at scale with an ELK Stack can be expensive, unstable, and relentlessly time-sucking. Now try ChaosSearch – a fully managed log analytics platform that delivers the Elasticsearch API you love, but with absolutely NO Elasticsearch under the hood! ChaosSearch leverages your own Amazon S3 as a data store, which means no data movement, no data retention limits and savings of up to 80% vs an ELK Stack. In fact with ChaosSearch, you just Store, Connect & Analyze to start experiencing insights at scale from ALL of your data (tell them Corey Quinn sent you)!
Dan Moore writes about his take on the most underappreciated AWS service.
What happens when you intentionally leak AWS keys.
The tale of how someone went from skeptic to enthusiast about AWS Lambda.
An analysis of Slack’s outage earlier this month shows that an AWS scale-up was partially to blame. This is what I mean when I talk about “knowing how your cloud provider breaks.” Different providers have different failure modes; knowing dragons like this lurk in the shadows is critical.
Honestly, “a collection of React components that help create intuitive, responsive, and accessible user experiences for web applications” coming from AWS is akin to an OSHA compliant safety framework from the creators of MTV’s Jackass.
If you’ve got an interesting job for this newsletter’s eminently employable subscribers, get in touch!
We’ve been benchmarking AWS vs Azure vs GCP for three years now. This year, we tested 54 machines, ran 1,000+ benchmarks, and questioned innumerable assumptions. We do it all for you, dear reader. To help you evaluate the clouds and their machines, and to help you choose the right configuration for your app.On many of the benchmarks, the margins were razor thin. Nevertheless, trends emerged. GCP’s throughput? Can’t be matched. Amazon’s network latency? Unbeatable. Intel chips? Stumbled. Wait, what now?Read the 2021 Cloud Report — or skim the highlights.
Software powers the world. LaunchDarkly is a feature management platform that empowers all teams to safely deliver and control software through feature flags. By separating code deployments from feature releases at scale, LaunchDarkly enables you to innovate faster, increase developer happiness, and drive DevOps transformation. To stay competitive, teams must adopt modern software engineering practices. LaunchDarkly enables teams to modernize faster. Intuit, GoPro, IBM, Atlassian, and thousands of other organizations rely on LaunchDarkly to pursue modern development and continuously deliver value. Visit us at launchdarkly.com to learn more.
Amazon CloudWatch Container Insights now available on AWS Graviton2 – AWS: Graviton2 is awesome! Also AWS: A year and change later some of our internal services are now aware Graviton2 exists.”
Amazon Detective enhances IP Address Analytics – The new issue is out, titled “Amazon Detective and the Case of Why is North Korea Talking To My Application.”
Amazon MSK now supports the ability to change the size or family of your Apache Kafka brokers – Credit where due: this also allows you to scale MSK down. An awful lot of AWS “elasticity” is only really designed to make things bigger.
AWS Certificate Manager Private Certificate Authority now supports additional certificate customization – The private CA adds some bells and whistles in an attempt to justify its $400 monthly price tag.
Using Route 53 Private Hosted Zones for Cross-account Multi-region Architectures – Well this is sure better than last week’s “replicate DynamoDB between accounts via backing it up to S3 and restoring it all the time.”
Amazon Redshift 2020 year in review – “We got our ass kicked by Snowflake” is the honest blog post on this topic, but apparently AWS enforces post standards that require articles to be longer than seven words.
Announcing the end of support for Python 2.7 in the AWS SDK for Python and AWS CLI v1 – This July, AWS will actually deprecate something–provided it doesn’t slip.
AWS Chalice adds support for the AWS CDK – Nobody has heard of Chalice in years, but that didn’t stop the CDK Evangelism Strike Team from paying them a visit in the dead of night.
How BT uses Amazon CloudWatch to monitor millions of devices – “With a metric crapton of money” presumably.
Customize 403 error pages from Amazon CloudFront Origin with Lambda@Edge – “Dynamically running code on every request” continues to intensely annoy me.
Measuring AWS Global Accelerator performance and analyzing results – “Here’s the series of measurements you should take that will justify using this product” is a heck of a blog post.
Stepping up for a truly open source Elasticsearch – tl;dr: Elastic relicensed their formerly open source offering, annoyed the entire Open Source community, AWS is now stepping in to offer a community open source Elasticsearch version, and Elastic is left more or less holding the bag. Whatever you think of this synopsis, it is critically important that you not email me.
Making retries safe with idempotent APIs – An AWS Builders Library entrant on how to make it safe for your users to spam the F5 key.
Gaining control of your cloud shouldn’t slow you down. That’s why cloudtamer.io gives you cloud control that won’t impact progress and innovation. You get a single solution to control cost, increase visibility and automation, and reduce risk. Think of it as your cloud governance multi-tool. One solution, many capabilities. Learn more and get 30 days free.
antiope is a decent attempt at an AWS Inventory and Compliance Framework.
A service/region availability tool that is actually usable.
… and that’s what happened Last Week in AWS.