AliyunContainerService/gpushare-scheduler-extender

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

1532 stars · 311 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A Kubernetes scheduler extender that enables multiple pods to share the same NVIDIA GPU device, increasing GPU utilization for inference and training workloads. It works with a companion device plugin to schedule GPU memory requests across pods on native Kubernetes clusters.

Use cases

• share a single GPU between multiple pods in Kubernetes
• increase GPU utilization for inference workloads
• schedule pods by GPU memory requests
• binpack pods onto the same GPU device
• run data science workloads on shared GPUs
• avoid dedicating a whole GPU to small tasks

When to choose

• you run Kubernetes with NVIDIA GPUs and pods need only a fraction of a GPU
• you want to improve GPU utilization without switching to a full GPU operator
• you use Alibaba Cloud Kubernetes and want a Helm-installable GPU sharing solution

When to avoid

• you need hard GPU isolation or QoS guarantees between tenants
• you require MIG-based partitioning or vendor-supported GPU sharing
• you need actively maintained software with recent releases and support

Facets

service · maturity maintenance

container-orchestration scheduling gpu-computing deployment cloud-computing infrastructure-as-code gpu-computing machine-learning go cloud gpu-sharing scheduler-extender device-plugin kubernetes-scheduler nvidia-gpu binpack resource-scheduling containers kubernetes docker linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="AliyunContainerService/gpushare-scheduler-extender")