Skip to Content
ChangelogMarch 2026 - V1.17.0

March 2026 - V1.17.0

✨✨✨ The most recent release brings a major new capability for faster image distribution, along with a set of usability improvements, performance enhancements, and important bug fixes. Check out what’s changed.

🚀 Highlights

Image Accelerator

CloudPilot AI now supports Image Accelerator, using P2P (peer-to-peer) image acceleration to optimize container image distribution across nodes. By reusing and sharing cached images between nodes instead of pulling every layer from the remote registry, it reduces image pull time, speeds up Pod startup, and lowers bandwidth usage.

  • Tries peer-to-peer image transfer first and falls back to the remote registry only when needed.
  • Runs on nodes provisioned by NodePools with Image Accelerator enabled.
  • Allows accelerator-enabled nodes to both consume cached images from peers and serve images to other nodes.
  • Reduces repeated registry pulls and improves cold-start performance when many Pods use the same image.
  • Provides cluster-wide and per-node acceleration metrics in the console for visibility and troubleshooting.
Image Accelerator workflow

Image Accelerator can be enabled through NodeClass and NodePool configuration, making it easy to roll out P2P acceleration to selected node groups while keeping adoption controlled and predictable. After setup, users can monitor total accelerated traffic, mirror request hits, and per-node serve/pull details directly in the console.

Image Accelerator page

⚙️ Enhancements

  • Support detecting errors in CloudPilot AI components and sending alerts, improving operational visibility.
  • Show user workload override information more clearly in both backend handling and product views.
  • Improve Universal optimization command generation for a smoother optimization workflow.
  • Add RecommendationPolicy and WorkloadOptimized filters for workload-related views.
  • Support enterprise discount configuration.
  • Improve usage query performance and update Workload Autoscaler related dependencies.

🛠️ Bug Fixes

  • Stop collecting failed or completed Pods to reduce unnecessary data noise.
  • Fix an issue where Pod eviction could get stuck under certain conditions.
  • Fix usage API request timeout issues.
  • Prevent errors when an invitation already exists.

These updates further improve CloudPilot AI’s overall reliability and user experience. For questions or support, join our Slack community 

Stay tuned for more updates! 🚀

Last updated on