Your submission was sent successfully! Close

Thank you for contacting us. A member of our team will be in touch shortly. Close

You have successfully unsubscribed! Close

Thank you for signing up for our newsletter!
In these regular emails you will find the latest updates about Ubuntu and upcoming events where you can meet our team.Close

Charmed Kubeflow 1.9 Beta is here: try it out

Tags: Kubeflow , MLOps

After releasing a new version of Ubuntu every six months for 20 years, it’s safe to say that we like keeping our traditions. Another of those traditions is our commitment to giving our Kubeflow users early access to the latest version – and that promise still stands. Kubeflow 1.9 is about to go out in a couple of weeks and that only means one thing: Canonical has just released its Charmed Kubeflow beta. Are you ready to try it out? 

If you can put some time aside, we’re looking for data scientists, ML engineers, MLOps experts, creators and AI enthusiasts to take Charmed Kubeflow 1.9 for a ride and share their feedback with us. You can really help us by:

  • Trying it out and letting us know how the experience goes
  • Asking us any questions about the product and how it works
  • Reporting bugs or any issues you have with Charmed Kubeflow 1.9 Beta (and beyond)
  • Giving us improvement suggestions for the product and portfolio  

What’s new in Kubeflow 1.9?

Kubeflow is now going through the CNCF process to graduate from the incubation program. This challenges the community to evolve quickly and work on different aspects of the projects:

  • Improving the MLOps platform’s security features
  • Adding new capabilities to the project
  • Centralising communication channels and growing the community

Security as a priority

This release has the first updates from the security working group. One of the key features that the group announced was about Network Policies, which control the traffic flow at the IP address or port level. They will be enabled as a second security layer for core services to give users a better network overview and segmentation in line with common enterprise security guidelines.

ML integrations as part of a growing ecosystem

Kubeflow is designed to work in partnership with other ML & data tools. The latest release brings news integrations with leading ML tools and libraries such as BentoML, used for inference, or Ray, for training LLMs. One long-standing bug that Charmed Kubeflow users reported was related to the access to MLflow when deployed alongside the MLOs platform. Charmed Kubeflow 1.9 will solve this issue and give users clear guidance on how to use it.

Community growth

As part of CNCF, the community aims to integrate better into the ecosystem and enable new contributors to the project. One of the changes that the upstream community just made was to move to the CNCF Slack channel. Join us there to get in touch with a vibrant community and learn more from some of the industry experts.

We’re going live; join our MLOps tech talk.

Speaking of traditions, you might already know that all our betas bring the product engineering team live for a tech talk. This time is no exception, and I’ll be joined by two new faces. Michal Hucko and Orfeas Kourkakis, Software Engineers at Canonical, are ready to talk tomorrow, 11 July 2024, at 5 PM CET to Kubeflow users about the platform, latest news and how the industry is being shaped. Join us live, and you will:

  • Learn about the latest release and how our distribution handles it
  • Discover the key features covered in Charmed Kubeflow 1.9, in upstream and beyond.
  • Understand the differences between the upstream release and Canonical’s Charmed Kubeflow.
  • Get answers to any other question, technical or not, you have about MLOps, open source or Canonical’s portfolio.

 Don’t wait any longer, and add the event to your calendar!

Charmed Kubeflow 1.9 is out. Try it now!

Are you already a Charmed Kubeflow user?

Your job is even easier since you will only have to upgrade to the latest version to try the 1.9 beta. We’ve already prepared a guide with all the steps you need to take. 

Please be mindful that this is not a stable version, so there is always a risk that something might go wrong. Save your work and proceed with caution. If you encounter any difficulties, Canonical’s MLOps team is here to hear your feedback and help you out. Since this is a Beta version, Canonical does not recommend running or upgrading it on any production environment.

Are you new to Charmed Kubeflow?

Now, I can tell you are a real adventurer. Welcome to the MLOps world! Starting with a beta release might result in a few more challenges for you, but it’ll give you the chance to share in the product development and really contribute to the open source world. For all the prerequisites, check out the getting started tutorial.

Shortly after you deploy and install MicroK8s and Juju, you will need to add the Kubeflow model and then make sure you have the latest version. Follow the instructions below to get this up and running:

juju deploy kubeflow –channel 1.9/beta –trust

Now, you can go back to the tutorial to finish the configuration of Charmed Kubeflow or read the documentation to learn more.

You tried it out – what do you think?

You are part of something really important for us. As with any other open source project, joining a beta gives you a glimpse of the latest innovations, and it also gives you the chance to shape the product. Let’s make Charmed Kubeflow 1.9 better together.

kubeflow logo

Run Kubeflow anywhere, easily

With Charmed Kubeflow, deployment and operations of Kubeflow are easy for any scenario.

Charmed Kubeflow is a collection of Python operators that define integration of the apps inside Kubeflow, like katib or pipelines-ui.

Use Kubeflow on-prem, desktop, edge, public cloud and multi-cloud.

Learn more about Charmed Kubeflow ›

kubeflow logo

What is Kubeflow?

Kubeflow makes deployments of Machine Learning workflows on Kubernetes simple, portable and scalable.

Kubeflow is the machine learning toolkit for Kubernetes. It extends Kubernetes ability to run independent and configurable steps, with machine learning specific frameworks and libraries.

Learn more about Kubeflow ›

kubeflow logo

Install Kubeflow

The Kubeflow project is dedicated to making deployments of machine learning workflows on Kubernetes simple, portable and scalable.

You can install Kubeflow on your workstation, local server or public cloud VM. It is easy to install with MicroK8s on any of these environments and can be scaled to high-availability.

Install Kubeflow ›

Newsletter signup

Get the latest Ubuntu news and updates in your inbox.

By submitting this form, I confirm that I have read and agree to Canonical's Privacy Policy.

Related posts

A deep dive into Kubeflow pipelines 

Widely adopted by both developers and organisations, Kubeflow is an MLOps platform that runs on Kubernetes and automates machine learning (ML) workloads. It...

Edge AI: what, why and how with open source

Edge AI is transforming the way that devices interact with data centres, challenging organisations to stay up to speed with the latest innovations. From...