Taming the Wild West of Research Computing - How Policies Saved Us a Thousand Headaches
CNCF [Cloud Native Computing Foundation] via YouTube
Start speaking a new language. It’s just 3 weeks away.
Earn a Michigan Engineering AI Certificate — Stay Ahead of the AI Revolution
Overview
Google, IBM & Meta Certificates – 40% Off
One plan covers every Professional Certificate on Coursera.
Unlock All Certificates
Learn how to implement effective governance and resource management for Kubernetes clusters in research computing environments through this 24-minute conference talk from CNCF. Discover practical strategies for controlling chaotic research workloads using policy-as-code approaches with Kyverno, Kueue, and Argo CD. Explore real-world solutions for preventing common issues like abandoned interactive GPU pods, misallocated CPU jobs on GPU nodes, and resource waste in bare-metal accelerated discovery clusters. Master techniques for enforcing fine-grained policies, implementing fair-share GPU scheduling, and automating governance without custom code or complex workarounds. Gain insights into applying GitOps discipline and smart policy design to maintain order in high-performance computing environments, whether managing research workloads or preventing general cluster chaos.
Syllabus
Taming the Wild West of Research Computing: How Policies Saved Us a Thousand... Alessandro Pomponio
Taught by
CNCF [Cloud Native Computing Foundation]