Apache YuniKorn
yunikorn.apache.org
- Free plan
- Web
- API
- Self-hosted
What Apache YuniKorn is
Apache YuniKorn is a resource scheduler for batch, data, and machine learning workloads running on Kubernetes. It provides an alternative to Kubernetes’ default scheduler and is described as transparent to existing Kubernetes applications. Its hierarchical queues let administrators set resource controls for tenants and reflect an organization’s structure. Scheduling options include gang scheduling, FIFO and Fair job ordering, resource fairness, reservations, and preemption. A web interface gives a centralized view of cluster capacity, utilization, and application information. YuniKorn exposes scheduling metrics through Prometheus, and project instructions cover monitoring with Prometheus and Grafana. The project recommends deployment to an existing Kubernetes cluster through its Helm chart, using Helm 3 or later. The software is free and open source under the Apache License 2.0. Releases include source code tarballs and prebuilt Docker images; source releases must be built before use. The project advises checking artifacts with GPG or SHA-512. Users can ask questions in Slack’s #yunikorn-user channel or on the user mailing list.
Who it suits
YuniKorn suits teams running batch, data, or machine learning workloads on Kubernetes that need tenant resource controls and scheduling policies. It also serves users and developers interested in using or contributing to the project.
Good to know, and what to watch
- Free software under the Apache License 2.0.
- Hierarchical queues provide fine-grained tenant controls.
- Scheduling includes gang scheduling, reservations, and preemption.
- Web UI summarizes capacity, utilization, and applications.
- Prometheus metrics support monitoring with Grafana.
- Source tarballs must be built before use.
- Deployment guidance targets an existing Kubernetes cluster.
- The project recommends Helm 3 or later.
Plans
Every plan and priceApache YuniKorn
Open source
The verdict
YuniKorn offers Kubernetes workload scheduling with queues, multiple scheduling policies, and a cluster overview. Teams should account for its deployment requirements and build step when using source releases.
What the maker publishes
- Deployment
- The project recommends deploying YuniKorn to an existing Kubernetes cluster using its Helm chart and Helm 3 or later. source
- Integrations
- YuniKorn exposes scheduling metrics via Prometheus, and the project provides instructions for monitoring with Prometheus and Grafana. source
- Intended users
- The community page identifies users and developers interested in using or contributing to YuniKorn as audiences for its community meetings. source
- Kubernetes scheduling
- YuniKorn is a Kubernetes-compatible alternative to the default Kubernetes scheduler and is described as transparent to existing Kubernetes applications. source
- License
- The software is licensed under the Apache License 2.0. source
- Purpose
- Apache YuniKorn is a resource scheduler for running batch, data, and machine learning workloads on Kubernetes. source
- Queues
- Hierarchical resource queues provide fine-grained resource controls for tenants and can map to an organization’s structure. source
- Release format
- YuniKorn is released as source code tarballs, and the project also publishes prebuilt Docker images. source
- Release verification
- The download page says release artifacts should be checked for tampering using GPG or SHA-512. source
- Scheduling features
- The scheduler supports gang scheduling, job ordering policies including FIFO and Fair, resource fairness, reservations, and preemption. source
- Security reporting
- The project asks people to report security issues to [email protected] and not disclose them publicly before working with the security team. source
- Support
- Users can ask for help in the YuniKorn Slack #yunikorn-user channel or use the user mailing list. source
- Web UI
- The web UI provides a centralized view of cluster capacity, utilization, and application information. source
Checked against the maker's pages on 7 Oct 2026.