Freedom251 › GPU Cluster Management Software

Volcano

Free plan APISelf-hosted

What Volcano is

Volcano is a free, self-hosted system for scheduling high-performance workloads on Kubernetes. It places native Kubernetes workloads and batch jobs under one scheduling system and supports frameworks including Spark, TensorFlow, PyTorch, Flink, Argo, MindSpore, PaddlePaddle, and Ray. Policies include gang, binpack, heterogeneous-device, capacity, fair-share, task-topology, and NUMA-aware scheduling. Multi-level queues provide quotas, resource inheritance, borrowing, reclaiming, and preemption. Volcano schedules across x86, Arm, Kunpeng, Ascend, GPU, and NPU resources, and can schedule jobs across clusters for larger pools and load balancing. It supports colocating online and offline workloads with dynamic resource overcommitment, CPU burst, and resource isolation. Logging, monitoring metrics, and a graphical dashboard are described for operations. Installation uses YAML, source code, or Helm on an existing Kubernetes cluster. Kubernetes 1.12 or later with CRD support is required. YAML installation supports x86_64 and arm64; documented one-click source installation is temporarily limited to x86_64. Volcano is a CNCF incubating project.

Who it suits

Volcano suits teams running Kubernetes batch or high-performance workloads that need queue controls, specialized scheduling policies, and multi-cluster resource management.

What it does well

Free and self-hosted
Supports Kubernetes batch and native workloads
Many scheduling policies
GPU and NPU resource support
Quota, borrowing, and preemption controls

What to know first

Requires an existing Kubernetes cluster
Kubernetes 1.12 or later is required
Source installation is temporarily x86_64-only

Limits

Volcano requires Kubernetes 1.12 or later with CRD support. YAML installation is available for x86_64 and arm64, while the documented one-click source installation is temporarily limited to x86_64.

Getting started

Open https://volcano.sh/.
Prepare an existing Kubernetes cluster running version 1.12 or later with CRD support.
Choose YAML, source code, or Helm as the installation route.
For YAML installation, use x86_64 or arm64; for documented source installation, use x86_64.

Verdict

Volcano provides extensive scheduling and resource controls for Kubernetes environments, including accelerators and cross-cluster jobs. Its self-hosted deployment requires Kubernetes administration and attention to the documented architecture limits.

Questions

How much does Volcano cost?

Volcano is free, with a free plan.

Where can Volcano run?

It is self-hosted on an existing Kubernetes cluster. The installation guide requires Kubernetes 1.12 or later with CRD support.

Which installation methods are available?

The listed options are YAML files, source code, and Helm.

Which architectures are supported?

YAML installation supports x86_64 and arm64. The documented one-click source installation is temporarily available only for x86_64.

Which workloads and frameworks does it support?

It schedules native Kubernetes workloads and batch jobs, and supports frameworks including Spark, TensorFlow, PyTorch, Flink, Argo, MindSpore, PaddlePaddle, and Ray.

How can users get project support?

The project directs users to the CNCF #volcano Slack channel, a mailing list, and GitHub issues and repositories.

Facts from Volcano's own pages

Architecture limit: The documented one-click installation from source is temporarily available only for x86_64; YAML installation supports x86_64 and arm64. · volcano.sh, 5 Oct 2026
Colocation: Volcano supports colocating online and offline workloads using unified scheduling, dynamic resource overcommitment, CPU burst, and resource isolation. · volcano.sh, 5 Oct 2026
Community support: The project directs users to the CNCF #volcano Slack channel, a mailing list, and GitHub issues and repositories. · volcano.sh, 5 Oct 2026
Deployment: Installation options include YAML files, source code, and Helm on an existing Kubernetes cluster. · volcano.sh, 5 Oct 2026
Frameworks: It supports frameworks including Spark, TensorFlow, PyTorch, Flink, Argo, MindSpore, PaddlePaddle, and Ray. · volcano.sh, 5 Oct 2026
Hardware: It supports scheduling across x86, Arm, Kunpeng, Ascend, GPU, and NPU resources. · volcano.sh, 5 Oct 2026
Monitoring: The introduction describes logging, monitoring metrics, and a dashboard for graphical operations. · volcano.sh, 5 Oct 2026
Multi-cluster: It supports cross-cluster job scheduling for larger resource pools and load balancing. · volcano.sh, 5 Oct 2026
Observability: The project describes logging, monitoring metrics, and a dashboard for graphical operations. · volcano.sh, 5 Oct 2026
Prerequisite: The installation guide lists Kubernetes 1.12 or later with CRD support as a prerequisite. · volcano.sh, 5 Oct 2026
Project status: Volcano is a Cloud Native Computing Foundation incubating project. · volcano.sh, 5 Oct 2026
Purpose: Volcano is a cloud native system for high-performance workloads and a container batch scheduling project. · volcano.sh, 5 Oct 2026
Queue management: Volcano supports multi-level queues, multidimensional resource quotas, resource inheritance, borrowing, reclaiming, and preemption. · volcano.sh, 5 Oct 2026
Resource management: It provides multi-level queues, resource quotas, borrowing, reclaiming, and preemption between queues. · volcano.sh, 5 Oct 2026
Scheduling policies: Its scheduling policies include gang, binpack, heterogeneous-device, capacity, fair-share, task-topology, and NUMA-aware scheduling. · volcano.sh, 5 Oct 2026
Supported architectures: The YAML installation method is available for x86_64 and arm64, while installation from source is listed as temporarily available only for x86_64. · volcano.sh, 5 Oct 2026
What it does: Volcano is a cloud-native system for scheduling high-performance workloads on Kubernetes. · volcano.sh, 5 Oct 2026
Workloads: It schedules Kubernetes workloads and batch jobs for frameworks including Spark, TensorFlow, PyTorch, Flink, Argo, MindSpore, PaddlePaddle, and Ray. · volcano.sh, 5 Oct 2026

Where Volcano stands on Freedom251

GPU Cluster Management Software#15 of 21