-
Notifications
You must be signed in to change notification settings - Fork 864
Open
Description
Some schedulers can schedule according to the GPU type. such as :
Volcano extends its support to include reporting different types of GPU resources within elastic queue capacities. NVIDIA’s default Device Plugin does not distinguish between GPU models, instead reporting all resources uniformly as nvidia.com/gpu. This limits AI training and inference tasks from selecting specific GPU models, such as A100 or T4, based on their particular needs. To address this, Volcano now supports reporting distinct GPU models at the Device Plugin level, working with the capacity plugin to enable more precise GPU resource sharing and recycling.
the template :
apiVersion: scheduling.volcano.sh/v1beta1
kind: Queue
metadata:
name: demo-queue
spec:
reclaimable: true
deserved: # set the deserved field.
cpu: 64
memeory: 128Gi
nvidia.com/a100: 40
nvidia.com/v100: 80
Metadata
Metadata
Assignees
Labels
No labels
Activity
upczsh commentedon May 8, 2025
Therefore, we can add a new gpu_type parameter in the kubernetes decorator to allow users to select the gpu type