stoneni

k8s上安装好插件，无法识别到集群GPU资源

1

k8s上安装好插件，pod调度成功，处于running状态，执行kubectl inspect gpushare获取GPU资源为0，调度一个推理服务提示：0/2 nodes are available: 2 Insufficient aliyun.com/gpu-mem. preemption: 0/2 nodes are available: 2 No preemption victims found for incoming pod

目前我用过的是K8s版本是1.24.2，是否支持？

Now a node has two RDMA physical network cards, each of which is virtualized by SRIVO with 4 VF. Now I create a Pod, and I need to apply for...

When a pod applies for 2 VF, only one net network interface can be seen inside the pod, and 2 are expected. What is the reason? Is there a limit...

### Ⅰ. Describe what this PR does Added end-to-end support for rdma devices, including device discovery, device registration, node resource update, scheduling, and allocation ### Ⅱ. Does this pull request...

### Summary This proposal provides a mechanism for the end-side dp to sense that the pod is being scheduled on the current node, get the device id according to the...

### Summary This proposal provides a mechanism for the end-side dp to sense that the pod is being scheduled on the current node, get the device id according to the...

The center side scheduling results and device allocation results make dp aware of lightweight methods

2

Since both the scheduling result and the assignment device are implemented in the scheduling plug-in, the scheduling result is marked in the pod Annotations. And dp is not aware of...

Koordinator does not currently support RDMA PF or VF (RDMA virtual network Card) resource management and scheduling allocation. But Koordinator seems to have supported RDMA scheduling, and also implemented allocation...