Use cGPU

Use scenarios

• Only dedicated Kubernetes clusters that contain GPU-accelerated nodes support the ack-cgpu component. Managed Kubernetes clusters that contain GPU-accelerated nodes do not support the ack-cgpu component.
• If you want to use cGPU in professional Kubernetes clusters, see Enable GPU sharing.

Procedure

1. Log on to a master node and run the following command to query the status of GPU sharing in your cluster.
   Note

   • For more information about how to log on to a master node, see Connect to a Linux instance by using password authentication or Connect to a Windows instance by using password authentication.
   • If you want to query the GPU sharing status of a cluster from an on-premises machine, you must install ack-cgpu and a GPU inspection tool. For more information, see Step 4: Install and use the GPU scheduling inspection tool.

   kubectl inspect cgpu

   Expected output:

   NAME                     IPADDRESS    GPU0(Allocated/Total)  GPU1(Allocated/Total)  GPU Memory(GiB)
   cn-beijing.192.168.0.4   192.168.0.4  0/7                    0/7                    0/14
   ---------------------------------------------------------------------
   Allocated/Total GPU Memory In Cluster:
   0/14 (0%)

   Note To query detailed information about GPU sharing, run the kubectl inspect cgpu -d command.
2. Deploy a sample application that has cGPU enabled and request 3 GiB of GPU memory for the application.

   apiVersion: apps/v1
   kind: StatefulSet
   metadata:
     name: binpack
     labels:
       app: binpack
   spec:
     replicas: 1
     serviceName: "binpack-1"
     podManagementPolicy: "Parallel"
     selector: # define how the deployment finds the pods it manages
       matchLabels:
         app: binpack-1
     template: # define the pods specifications
       metadata:
         labels:
           app: binpack-1
       spec:
         containers:
         - name: binpack-1
           image: registry.cn-beijing.aliyuncs.com/ai-samples/gpushare-sample:tensorflow-1.5
           command:
             - python
             - cgpu/main.py
           resources:
             limits:
               # GiB
               aliyun.com/gpu-mem: 3

   Note aliyun.com/gpu-mem: specifies the amount of GPU memory requested by the application.
3. Run the following command to query the memory usage of the GPU:

   kubectl inspect cgpu

   Expected output:

   NAME                      IPADDRESS      GPU0(Allocated/Total)  GPU Memory(GiB)
   cn-beijing.192.168.1.105  192.168.1.105  3/14                   3/14
   ---------------------------------------------------------------------
   Allocated/Total GPU Memory In Cluster:
   3/14 (21%)

   The output shows that the total GPU memory of the cn-beijing.192.168.1.105 node is 14 GiB and 3 GiB of GPU memory has been allocated.