← Scheduling & Orchestration
Core
Slurm for AI Clusters
Slurm allocates resources to batch jobs and launches their tasks. GPU requests, task placement and queue policy work together to determine when a training job runs.
This concept is included with a free account. Sign in to continue reading.
free account · no card needed
RELATED CONCEPTS
LESSONS THAT TEACH THIS
PRACTICE THIS IN REAL QUESTIONS
Kubernetes, Slurm & GPU SchedulingExplain how a Slurm job actually gets its GPUs: GRES, partitions, cgroups, and the sbatch that lands 8 GPUs on one node.→Kubernetes, Slurm & GPU SchedulingSlurm or Kubernetes for a 2,000-GPU training cluster? Make the case, and tell me what you lose either way.→AI Infrastructure System DesignDesign a training cluster for 10,000 GPUs: pods, fabric, storage, failure domains and the scheduler. Justify every number.→Kubernetes, Slurm & GPU SchedulingWalk me through what happens when a pod asks Kubernetes for four GPUs, from the manifest to the container seeing them.→AI Infrastructure System DesignDesign a training cluster for a one-trillion-parameter MoE. Size it, choose the parallel layout, and map it onto the fabric.→AI Infrastructure System DesignWalk me through an inference platform for a hosted LLM. What are the pieces, and what does each one do?→