Ubuntu software developer Canonical Ltd. today launched its machine learning operations toolkit Charmed Kubeflow on Amazon Web Services Inc.’s cloud marketplace.

Charmed Kubeflow is available as a software application on AWS, making it simpler for companies to launch and manage machine learning workloads. The software is an enterprise-grade version of Kubeflow, an open-source MLOps toolkit that’s designed to work with Kubernetes, the popular application container orchestration software. It provides various tools that simplify the task of running artificial intelligence on Kubernetes.

By bringing Charmed Kubeflow to AWS, Canonical said it’s providing companies with a quick path to machine learning operations experimentation. It comes at a time when a growing number of organizations are showing an outsized interest in artificial intelligence and machine learnings. According to IBM Corp.’s Global AI Adoption Index, 35% of enterprises reported that they’re adopting AI in 2022. With the rise of generative AI projects such as ChatGPT, interest in the technology is accelerating rapidly.

Canonical said Charmed Kubeflow on AWS is intended for companies looking to kickstart their AI and machine learning initiatives because it’s easy to deploy and provides as much computing power as they need to experiment without constraints. Using the platform, it’s possible to set up a new MLOps environment in a matter of minutes, Canonical said.

Charmed Kubeflow works by automating machine learning workflows to create a reliable application layer for model development, iteration and production deployment. It also provides full visibility into those workloads, so teams can assess any challenges and accurately plan their infrastructure expansion requirements.

Once the experimentation phase is complete, users can deploy their models on end devices. At the same time, Charmed Kubeflow will provide protection against cyberattacks with regular scanning, patching and upgrades to the latest version of the machine learning libraries being used. For production-grade deployments, users can shift artifacts from the Charmed Kubeflow appliance to a supported deployment on AWS or in their own data center.

Aaron Whitehouse, Canonical’s senior public cloud enablement director, said Charmed Kubeflow is an ideal platform for companies that are looking to explore machine learning for the first time. “The Charmed Kubeflow appliance on AWS gives companies a great way to test out machine learning possibilities quickly and easily, with a clear pathway to a scalable hybrid/multicloud deployment if those pilot projects are successful,” he said.

Charmed Kubeflow on AWS can be accessed via the AWS Marketplace now, with a fully managed option available for organizations requiring infrastructure support.

Image: Canonical