HAMi vs Volcano
HAMi
project-hami.io
Volcano
volcano.shSide by side
| Figure | HAMi | Volcano |
|---|---|---|
| Price from | Free | — |
| Free plan | Yes | — |
| Free trial | — | — |
| Open source | — | — |
| Linux | Yes | — |
| API | — | Yes |
| Self-hosted | Yes | Yes |
| Plans published | 1 | 0 |
| Spec score · GPU Cluster Management Software | 8.3 (#5) | 6.5 (#19) |
| Community support | The project links to community support through Discord and Slack (#hami-dev). | The project directs users to the CNCF #volcano Slack channel, a mailing list, and GitHub issues and repositories. |
| Deployment | The quick start installs HAMi with Helm and requires Kubernetes, Helm, kubectl, and installation permissions. | Installation options include YAML files, source code, and Helm on an existing Kubernetes cluster. |
| Monitoring | HAMi provides allocation counts and spread plus real-time GPU memory and core utilization visibility. | The introduction describes logging, monitoring metrics, and a dashboard for graphical operations. |
| Project status | HAMi is a CNCF Incubating project. | Volcano is a Cloud Native Computing Foundation incubating project. |
| Purpose | HAMi is open-source, cloud-native GPU virtualization middleware for sharing, isolating, and scheduling heterogeneous accelerators on Kubernetes. | Volcano is a cloud native system for high-performance workloads and a container batch scheduling project. |
| Workloads | The site identifies LLM, machine-learning, and HPC workloads as use cases. | It schedules Kubernetes workloads and batch jobs for frameworks including Spark, TensorFlow, PyTorch, Flink, Argo, MindSpore, PaddlePaddle, and Ray. |
Plans
In their own words
HAMi
HAMi is open-source middleware for sharing and scheduling GPUs and other accelerators among AI workloads on Kubernetes. Workloads can request GPU memory and core limits through Kubernetes resource controls. Scheduling options include binpack, spread and numa-first, with binpack and spread available at node or GPU scope. The supported-device list spans NVIDIA, AMD, AWS Neuron, Huawei Ascend, Cambricon, Enflame, Hygon, Iluvatar, Kunlunxin, MetaX, Moore Threads, Vastai and Biren; isolation and partitioning capabilities vary by device. HAMi works with Kubernetes APIs, DRA and CDI, and documents scheduling integrations including Volcano, Kueue, Koordinator and NVIDIA KAI Scheduler. Its node metrics endpoint exposes device memory and core utilization for Prometheus. Classic HAMi requires Kubernetes v1.23 or later. HAMi-DRA requires Kubernetes v1.34 or later, the DRA Consumable Capacity feature gate, CDI and NVIDIA driver 440 or later. NVIDIA limits use user-space library interception, which does not cover applications that bypass CUDA. Enforcement is described as soft and best-effort; the FAQ recommends MIG when hardware-enforced isolation is required.