distribution-spec VS swarmkit

Back in 2013, a little company called Docker made it really easy to start using containers to package up applications. A big key to their success was the OCI (you can learn about that here), an industry wide initiative to have standards around how we package up our applications. Because of OCI standards, we have hundreds (maybe thousands?) of tools that can be combined to manage and deploy applications. So why aren’t we using this for packaging up Notebooks and AI models as well? It would make deploying, sharing, and managing our models easier for everyone involved.

Kubernetes on the backend used to utilize docker for much of its container runtime solutions. One of the modular features of Kubernetes is the ability to utilize a Container Runtime Interface or CRI. The problem was that Docker didn't really meet the spec properly and they had to maintain a shim to translate properly. Instead users could utilize the popular containerd or cri-o runtimes. These follow the Open Container Initiative or OCI's guidelines on container formats.

OCI stands for Open Container Initiative, and its goal as an organization is to define a specification for container formats and runtime.

https://opencontainers.org/

Here is Containerfile from the repo: https://github.com/ublue-os/bazzite/blob/main/Containerfile

apko allows us to build OCI container images from .apk packages.

Flagger's load testing service can be installed via a Kustomization resource based on manifests packaged as an artifact in an Open Container Initiative (OCI) registry

CRI-O is a lightweight container runtime for Kubernetes. It is an implementation of Kubernetes CRI to use Open Container Initiative (OCI) compatible runtimes for running pods. It supports runc and Kata Containers as the container runtimes, but any OCI-compatible runtime can be integrated.

So somebody else created runj (runj is an experimental, proof-of-concept OCI-compatible runtime for FreeBSD jails.) https://github.com/samuelkarp/runj

It's supported, though the development pace is rather slow. The are new features (the latest release brought support for the Container Storage Interface) and bugs and issues are worked on by the few maintainers the project has. You can check the repo here: https://github.com/moby/swarmkit

So the thing is support for Swarm was delegated to Mirantis, https://www.mirantis.com/blog/mirantis-will-continue-to-support-and-develop-docker-swarm/ since it was delegated very little was done to move forward swarm _> https://github.com/moby/swarmkit/commits/master , docker swarm itself (docker the company) is deprecated https://github.com/docker-archive/classicswarm . I think because there's no way to monetize swarm there's no real push for new features hence my assumption of deprecation.

Feature request: Enable "--device" in "stack deploy" for Confidential Computing

This doesn't look that abandoned? https://github.com/moby/swarmkit

Or are you talking about swarm the product (versus docker swarm mode)

I don’t feel a momentum in Docker Swarm development, seeing pull requests like #3072 sitting idle for half a year, originally from 2016.

privileged not being supported is problematic, but there are PRs for it on swarmkit, see https://github.com/moby/swarmkit/pull/3072 - let's see if that gets in.

swarm ha placement

It is already in a different repo: moby/swarmkit

Docker Swarm is dead. It should not be used for new projects because the competitive products are much better and more optimal. Check their repo bro - last commit 3months ago. https://github.com/docker/swarmkit https://www.reddit.com/r/docker/comments/ituysl/why_people_are_saying_docker_swarm_is_in_life/ Why would you start a project with a tool which got no future? Let's then use Silverlight or any shit like that.

Mostly it just needs resource defaults merged and some abstract type support so load balancer config doesn't have to be entirely by labels and I would never shift from it.