Cloud Native News - CNN21/48

Editorial

• Will Nix Overtake Docker?
  Read about the side-by-side comparison of Nix and Docker. When does it make sense to use which? And where are both tools currently going? As you can build Docker images with Nix, and with Nix some problems of the Dockerfile approach are solved, we are looking into an exciting alternative.
• 3 Reasons to Use Kubernetes Operators (and 2 Reasons Not To)
  Since Pixie is a complex application performance monitoring tool, the Pixie team believed the benefits of running an operator-based deployment heavily outweighed the downsides. Here is why...
• Benchmarking Linkerd and Istio: 2021 Redux
  Earlier this year, Linkerd published a Linkerd vs Istio benchmark, comparing the performance and resource consumption of the two service meshes on a simple microservice application under various levels of load. Linkerd was dramatically faster than Istio while consuming an order of magnitude less data plane memory and CPU. With the recent release of the authorization policy in Linkerd 2.11, the team wanted to reevaluate Linkerd’s performance.

Tutorials

• Anti-Patterns When Building Container Images
  An excellent summary of anti-pattern for container images, each pattern itself may not be that worse, but combined, they screw your images. This list also includes recommendations on avoiding or not applying these patterns.
• OpenFaaS - Run Containerized Functions On Your Own Terms
  "How to turn Kubernetes cluster into FaaS platform? How can standalone VM with nothing but containerd installed compete with full-blow container orchestrator?"
• How to Develop a Custom Provider in Terraform
  Terraform supports a brought field of platforms, but sometimes there is no provider available: e.g., for your private internal cloud or missing extensions. Here is how to get started with developing your own one.
• EdgeX Foundry on K3s - the Initiation
  Part 2 of EdgeX on K3s series demonstrates - how a particular sensor can be attached in a simplified way & how the sensor data can be processed using EdgeX.
• How to use dig
  A short tutorial on how to use dig to make DNS queries.

Community

• Try eBPF-powered Cilium Service Mesh - join the beta program!
  Cilium is bringing the power of eBPF to the world of Service Mesh, and Cilium invites you to be amongst the first to try it out. Join the beta program!
• DigitalOcean Kubernetes Challenge
  Whether you’re new to Kubernetes or have been running production clusters for years, DigitalOcean’s Kubernetes Challenge will give you an opportunity to level up your Kubernetes skill set before the end of the year.
• Knative Has Applied to Become a CNCF Incubating Project
  Google just announced its intention to donate Knative to the CNCF incubating project.
• News from AWS re:Invent – Docker Official Images on Amazon ECR Public
  In partnership with Amazon, Docker announced that Docker Official Images are now available on AWS ECR Public. Having them available on ECR Public, in addition to Docker Hub, makes it easier for Amazon customers to use these images conveniently and securely and gives developers the flexibility to download Docker Official Images from their choice of the registry.
• Kubernetes 1.23 – What’s new?
  We are very close to the next Kubernetes release v1.23, it's just a question of hours. The latest major release brings some very exciting features: kubectl events, OpenAPI V3, multiple deprecations, IPv4/6 dual-stack goes GA and so on

Tools

• Introducing Karpenter – An Open-Source High-Performance Kubernetes Cluster Autoscaler
  AWS announced that Karpenter is ready for production. Karpenter is an open-source, flexible, high-performance Kubernetes cluster autoscaler built with AWS. It helps improve your application availability and cluster efficiency by rapidly launching right-sized compute resources in response to changing application load. Karpenter also provides just-in-time compute resources to meet your application’s needs.

Other

• Demystifying kube-proxy
  "This blog post explains how the kube-proxy component of Kubernetes works internally. It goes through various laters of networking abstractions in a top-down approach, ultimately leading to kube-proxy."

Photo by Patrick Federi on Unsplash