The Sched app allows you to build your schedule but is not a substitute for your event registration. You must be registered for KubeCon + CloudNativeCon Europe 2025 to participate in the sessions. If you have not registered but would like to join us, please go to the event registration page to purchase a registration.
Please note: This schedule is automatically displayed in British Summer Time (BST) (UTC +1). To see the schedule in your preferred timezone, please select from the drop-down menu to the right, above "Filter by Date." The schedule is subject to change and session seating is available on a first-come, first-served basis.
What if you had to update 100,000 objects every second? How do we avoid choking up the API server with such a load? Is directly modifying etcd objects even a viable option? In Kueue, we asked those questions a lot, and honestly, using standard CustomResourceDefinitions alone couldn’t keep up with the demands of this scale. In this session we will showcase how we combined the best of both worlds: CRDs and the good old K8s API Aggregation Layer to handle the dynamic positioning of up to 100,000 jobs in the queue.
Attendees will also gain valuable insights into system design decisions when dealing with such a scale and practical lessons for addressing similar challenges. We will delve into the pros and cons of both CRDs and the Aggregation Layer, providing clear guidance on when and why to use each. Finally, attendees will see how the K8s API Aggregation Layer extension works in practice and receive hands-on knowledge on choosing the right tool for the job.
Patryk is a Software Engineer at Google working on Kueue, an open-source SIG project. Member of the Kubernetes Batch Working Group, always willing to collaborate and share his knowledge.
