Take a run through some of the thought-provoking content I came across today. There’s some useful material for designing SLOs, soliciting feedback, and using modern dev workstations.
[blog] Adopting SRE: Standardizing your SLO design process. Useful content here, but the best thing is a link to a template you can download and use to define service level objectives for your workload. Adopting SLOs is a powerful way to create a common language around what’s important for your service.
[article] How Leaders Can Get the Feedback They Need to Grow. this post has wonderful advice for soliciting and processing feedback, even the tough feedback we don’t love hearing.
[blog] What Google Cloud VMware Engine can do for you: Three customers talk TCO. Lifting and shifting your hypervisor to the cloud doesn’t FEEL to me like the best long term strategy. But there are lots of reasons it works in the short term, and I liked this post which called out one or two benefits I hadn’t really thought of before.
[article] The Great Lambda Migration to Kubernetes Jobs—A Journey in Three Parts. I think I see more folks go the other direction (from VMs or Kubernetes into something serverless), but this happens too. If you use a container-based set of runtimes (hint, GKE, Cloud Run, and even Compute Engine) this migration isn’t that big of a deal.
[blog] Building a Media Understanding Platform for ML Innovations. The engineering team at Netflix built a platform to help their teams find just the right dialog or media from their movies to use in promotional videos.
[blog] Reducing GKE Log Ingestion. This solves a fairly specific problem, but I like reading blog posts that walk you through a solution after explaining the problem area.
[youtube-video] How to create a serverless Cloud Workstation. Here’s a five minute video that digs into online dev environments. Whether you code every day, or open an IDE once a week, this model has a lot of benefits.
[blog] AWS’s Anti-Competitive Move Hidden in Plain Sight. Interesting. I honestly had to look and see if we’re any better. Appears that we charge a single price for cross-region replicas, whether you’re using VMs or managed DBs.
[repo] GitHub OSPO. Building out an open source program office internally to support your consumption and participation in open source communities? GitHub just published some useful guidance.
##
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 