Volcano
volcano.sh
- Free plan
- API
- Self-hosted
What Volcano is
Volcano is a free, self-hosted system for scheduling high-performance workloads on Kubernetes. It combines native Kubernetes workloads and batch jobs in one scheduling system, and supports Spark, TensorFlow, PyTorch, Flink, Argo, MindSpore, PaddlePaddle, and Ray. Scheduling policies include gang, binpack, heterogeneous-device, capacity, fair-share, task-topology, and NUMA-aware approaches. Multi-level queues, multidimensional quotas, resource inheritance, borrowing, reclaiming, and preemption organize resources between queues. Volcano can schedule across x86, Arm, Kunpeng, Ascend, GPU, and NPU resources, and supports cross-cluster scheduling for larger pools and load balancing. It also supports colocating online and offline workloads with unified scheduling, 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 supports x86_64 and arm64; one-click source installation is temporarily x86_64-only.
Who it suits
Volcano suits teams running batch and high-performance workloads on Kubernetes, especially where queue controls, specialized hardware, or multiple clusters matter. It is also aimed at organizations colocating online and offline workloads.
Good to know, and what to watch
- Free plan
- Supports major machine-learning and data frameworks
- Many scheduling policies
- Cross-cluster scheduling
- GPU and NPU resource support
- Requires Kubernetes 1.12 or later
- Source installation temporarily supports x86_64 only
- Self-hosted deployment requires an existing cluster
The verdict
Volcano brings broad workload, queue, hardware, and cluster scheduling controls to Kubernetes. Its documented installation and architecture requirements should be checked before deployment.
What the maker publishes
- Architecture limit
- The documented one-click installation from source is temporarily available only for x86_64; YAML installation supports x86_64 and arm64. source
- Colocation
- Volcano supports colocating online and offline workloads using unified scheduling, dynamic resource overcommitment, CPU burst, and resource isolation. source
- Colocation
- It can schedule online and offline workloads in the same cluster, using resource oversubscription and isolation to improve utilization while prioritizing online workload quality of service. source
- Community support
- The project directs users to the CNCF #volcano Slack channel, a mailing list, and GitHub issues and repositories. source
- Deployment
- Installation options include YAML files, source code, and Helm on an existing Kubernetes cluster. source
- Frameworks
- It supports frameworks including Spark, TensorFlow, PyTorch, Flink, Argo, MindSpore, PaddlePaddle, and Ray. source
- Hardware
- It supports scheduling across x86, Arm, Kunpeng, Ascend, GPU, and NPU resources. source
- Monitoring
- The introduction describes logging, monitoring metrics, and a dashboard for graphical operations. source
- Multi-cluster
- It supports cross-cluster job scheduling for larger resource pools and load balancing. source
- Multi-cluster
- The Volcano Global sub-project supports cross-cluster task distribution, resource management, and priority control. source
- Observability
- The project describes logging, monitoring metrics, and a dashboard for graphical operations. source
- Prerequisite
- The installation guide lists Kubernetes 1.12 or later with CRD support as a prerequisite. source
- Project status
- Volcano is a Cloud Native Computing Foundation incubating project. source
- Purpose
- Volcano is a cloud native system for high-performance workloads and a container batch scheduling project. source
- Queue management
- Volcano supports multi-level queues, multidimensional resource quotas, resource inheritance, borrowing, reclaiming, and preemption. source
- Resource management
- It provides multi-level queues, resource quotas, borrowing, reclaiming, and preemption between queues. source
- Scheduling policies
- Its scheduling policies include gang, binpack, heterogeneous-device, capacity, fair-share, task-topology, and NUMA-aware scheduling. source
- 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. source
- What it does
- Volcano is a cloud-native system for scheduling high-performance workloads on Kubernetes. source
- Workloads
- It schedules native Kubernetes workloads and batch jobs through a unified scheduling system. source
- Workloads
- It schedules Kubernetes workloads and batch jobs for frameworks including Spark, TensorFlow, PyTorch, Flink, Argo, MindSpore, PaddlePaddle, and Ray. source
Checked against the maker's pages on 5 Oct 2026.