Best of Both Worlds: Integrating Slurm with Kubernetes in a Kubernetes Native Way
CNCF [Cloud Native Computing Foundation] via YouTube
Overview
Coursera Spring Sale
40% Off Coursera Plus Annual!
Grab it
Learn about K-Foundry, an open-source custom controller for KCP, in this 34-minute conference talk that demonstrates how to effectively bridge Slurm and Kubernetes orchestration systems. Explore how organizations can overcome the challenge of choosing between Slurm's robust job scheduling capabilities and Kubernetes' dynamic orchestration for AI/ML workloads. Discover how K-Foundry translates Kubernetes jobs to Slurm jobs and exposes Slurm nodes and cluster information as Kubernetes Custom Resource Definitions (CRDs), enabling seamless integration between cloud and HPC environments. Watch a live demonstration showcasing how this unified approach simplifies cluster administration through a common API, optimizes resource management, and enhances performance for large-scale AI and LLM tasks.
Syllabus
Best of Both Worlds: Integrating Slurm with Kubernetes in a Kubernetes... E.A. Gutierrez, A. Beltre
Taught by
CNCF [Cloud Native Computing Foundation]