Scheduling Classes

Mon, 01 Jan 0001 00:00:00 +0000

SchedulingClass is a cluster-scoped custom resource that lets administrators define placement policies for tenant workloads. When a tenant is assigned a scheduling class, all of its pods are automatically routed to the Cozystack custom scheduler, which merges the class-defined constraints with any constraints already present on the pod.

This allows platform operators to pin tenants to specific data centers, availability zones, or node groups — without modifying individual application charts.

Running Containerized GPU Workloads

Mon, 01 Jan 0001 00:00:00 +0000

This page covers running GPU workloads in regular Kubernetes pods (CUDA, ML training, inference) on Cozystack management cluster nodes. It targets the typical Linux GPU node shape — apt-installed (or dnf-installed) NVIDIA driver plus nvidia-container-toolkit — and uses the container variant of the cozystack.gpu-operator package.

If instead you want to pass whole GPUs to KubeVirt VMs, see GPU Passthrough and GPU Sharing with HAMi (HAMi is for fractional sharing in tenant Kubernetes clusters and is orthogonal to this variant; you can stack HAMi on top once the container variant is up).

Cluster Configuration and Management Guide on Cozystack

Scheduling Classes

Running Containerized GPU Workloads