8 chips not available

[steady storm]

I'm setting up a training run with some H100s. I downloaded the data and everything on one chip to then stop and start again with 8 when ready to train. But now when i try to start the pod again, there is only the option for using 3 chips, not the whole 8. Is this because of high demand at the moment or smtn?

[cyan gorge]

To help others find answers, you can mark your question as solved via Right click solution message -> Apps -> ✅ Mark Solution

[granite cliff]

yes, usually it is because of the high demand for the given GPU within the specific hosting site / node where your pod is located. If you are using network storage and not pod volume, just terminate your pod and deploy it in a site with more H100 available. If using pod volume, you have to wait and try later.

[steady storm]

Alr, thanks!