Collisions in the Dark: Illuminating the 95% of Kubeflow You Can't... Amine Lahouel & Laura Llinares
About this talk
This talk, led by Amine Lahouel and Laura Llinares from CERN, explores the critical need for effective monitoring within Kubeflow, especially as the use of their machine learning platform scales. The speakers share their experience in developing a comprehensive observability stack designed to provide actionable insights into performance, resource consumption, and overall system health. They demonstrate how they utilized established monitoring tools and created essential components to enhance visibility into the workload metrics, catering to both researchers who need detailed data on GPU utilization and Platform Managers who prefer broader usage trends and cost analysis.
More from this event
See all 436 talks →
Best of KubeCon + CloudNativeCon Amsterdam 2026
2:17
The Quiet Work of Forever: Sustaining Open Source Communities - O. Hope Amaechi-Okorie, JSON Schema
26:24
Evolving KServe: The Unified Model Inference Platform for Both Predictive and... F. Spolti & J. Lee
32:40
Preventing S3 Cost Storms: Applying Cortex’s Efficiency Lessons to I/O-Heav... A. Fishman-Lichterman
5:32