Announcing DKube™ 3.0

DKube 3.0 is now in production, and includes important new capabilities:

  • DKube Monitor, a state-of-the-art model monitoring capability, has been integrated into the leadership DKube set of end-to-end MLOps features
  • DKube Monitor can be licensed as part of the comprehensive DKube Full Suite, or licensed separately and used to monitor models created outside of DKube
  • DKube has been upgraded to incorporate Kubeflow 1.3, the newest version of the leadership standard
  • DKube supports A100 GPUs with multi-instance support, and runs out of box on A100-based DGX systems

DKube Monitor enables the production engineer to ensure that a deployed model continues to provide acceptable inference results as the input data or business goals change over time. Alerts can be set up based on measurement tolerances, and the results are viewed through an intuitive dashboard. Deviations can be quickly identified and resolved through retraining and redeployment.

This capability is as important as the initial model development, since the performance of deployed models will generally degrade over time, and the results need to be identified and improved before they impact business outcomes.

These new features add to the leadership capabilities DKube users have come to expect:

  • Based on best-in-class community standards, including Kubeflow and MLFlow
  • Operation on-prem or in the cloud, with the same look, feel, & workflow on both, and the ability to easily migrate between them
  • An intuitive, integrated end-to-end workflow, from Feature Engineering to Deployment
  • Automatic lineage and tracking to enable retraining by cloning a run and making the necessary incremental improvements
  • Automation of your workflow through CI/CD & Kubeflow Pipelines

Written by
Team DKube

The time to put your AI model to work is now

There's a faster way to go from research to application. Find out how an MLOps workflow can benefit your teams.

Schedule a Demo