Today’s reading/watching list included a good set of product updates, thought pieces, and a wildly catchy video tune. Enjoy!
[blog] Promoting pre-prod to production in Cloud Run with Google Cloud Deploy. With serverless platforms, and cloud in general, there isn’t always the traditional idea of “distinct environments” when deploying software. This post looks at one strategy.
[article] Should Every Incident Get a Retrospective? How do you choose when to do a retrospective for an incident? This article looks at criteria that may come into play.
[blog] Are Serverless Services Worth It? I have plenty of opinions about what the right solutions are for abstract problems. But “right” isn’t always “best” given the context, as this post points out.
[youtube-video] The Software Supply Chain Song (I’m Goin’ To Production). I bet you’ll be humming this song four a few hours after watching this.
[blog] EA Principles Addendum: Is Composability a good idea? How do you avoid over-engineering in the pursuit of architectural ideals like “composability”? Brian explores that in this post.
[blog] Monitor the health of your VM fleets in the Compute Engine console. Yes, serverless is great. SaaS is terrific. But you’re probably using a lot of VMs. Read this post about new fleet-level observability for VMs in Google Cloud.
[article] Kubernetes Alternatives to Spring Java Framework. When Spring Cloud first came out, it offered a unique way for Java devs to implement powerful microservices patterns. It still does. Now, platforms like Kubernetes and Istio have built-in support for many of the same patterns, and make them available to apps in any programming language.
[blog] Do the numbers: How AI is helping revolutionize accounting. AI is coming for all the things. Even accountants can be bad-asses with AI now, apparently.
[article] Orgs Shift Toward Centralizing Automation Initiatives. Centralize automation through tool consolidation, or maybe by incorporating platform engineering concepts? This article reviews new survey data about automation sprawl.
[blog] Autopilot is now GKE’s default mode of operation — here’s what that means for you. Like with any platform of reasonable size, there’s a lot to know when using Kubernetes. But choose services that let you know less. GKE Autopilot takes care of most all day-2 activities, and is now the default and recommended choice in the UI.
[report] Perspectives on Security for the Board. The Google Cloud Cybersecurity Action Team published a report on issues that Boards should know about.
##
Want to get this update sent to you every day? Subscribe to my RSS feed or subscribe via email below:
Richard Seroter's Architecture Musings 
One thought