Operations | Monitoring | ITSM | DevOps | Cloud

Canonical

Open Source MLOps on AWS

With the rise of generative AI, enterprises are growing their AI budgets, looking for options to quickly set up the infrastructure and run the entire machine learning cycle. Cloud providers like AWS are often preferred to kick-start AI/ML projects as they offer the computing power to experiment without long-term commitments. Starting on the cloud takes away the burden of computing power, reducing start-up time and cost and allowing teams to iterate more quickly.

Tuning a real-time kernel

This blog expands our technical deep-dive into a real-time kernel. You will need to be familiar with a real-time kernel to understand the tuning concepts in this blog. If you are starting from scratch and need to revisit the basics of preemption and a real-time system, watch this introductory webinar. If you are interested in the primary test suites for real-time Ubuntu, an explanation of the components and processes involved, head over to the first part of this mini-series.

Kubeflow vs MLFlow

Learn the main differences between the MLOps tools of choice: Kubeflow and MLFlow Started by Google a couple of years ago, Kubeflow is an end-to-end MLOps platform for AI at scale. Canonical has its own distribution, Charmed Kubeflow, which addresses the entire machine-learning lifecycle. Charmed Kubeflow is a suite of tools, such as Notebooks for training, Pipeline for automation, Katib for hyperparameter tuning or KServe for model serving and more. Charmed Kubeflow benefits from a wide range of integrations with other tools such as MLFlow, Spark, Grafana or Prometheus.

Securing open source software with Platform One and Canonical

Our own Devin Breen and Mark Lewis discussed Securing Open Source Software with the Chairman of Iron Bank at USAF Platform One Zachary Burke at AWS Summit Washington, DC. The topic includes: Securing Open Source Software, Secure Minimal Containers, and Software Security Scanning.

How we improved testing Ubuntu on WSL - and how you can too!

As the popularity of Windows Subsystem for Linux increases, the Ubuntu development team is committed to delivering a first class experience for Linux developers on Windows. To achieve this we’ve made improvements to our automated testing workflows via the creation of WSL-specific GitHub actions. In this post, Ubuntu WSL engineer Eduard Gómez Escandell talks us through the motivation for this approach and how you can implement these actions for your own CI workflows.

Canonical solutions reduce SmartNIC time-to-market and drive efficiency in enterprise data centres

Data centre efficiency is a central cost factor in enterprise IT environments. So, in the face of rising energy costs and increasingly resource-intensive workloads, it is more important than ever for businesses to seek out greater optimisation. Traditionally, CPUs have been used for the majority of data centre workloads, including network related tasks.